Processing method and device for evaluating perceptual response performance of vehicle

Abstract

The embodiment of the invention relates to a processing method and device for evaluating the perception response performance of a vehicle, and the method comprises the steps: initializing two empty queues and recording the two empty queues as corresponding vehicle / perception data queues, and initializing fourteen empty abnormal queues to form corresponding fourteen types of abnormal queues when the evaluation of the vehicle is started; in a vehicle evaluation process, data acquisition is carried out on real-time motion, positioning and road information output by a vehicle-mounted system to generate corresponding vehicle data records, the vehicle data records are added into a vehicle data queue, data acquisition is carried out on target sensing data output by the vehicle-mounted system to generate corresponding sensing data records, and the sensing data records are added into a sensing data queue; according to the vehicle data queue, the sensing data queue and the fourteen types of exception queues, fourteen types of sensing response exception recognition processing are carried out; and after the evaluation of the vehicle is finished, performing comprehensive scoring according to the fourteen types of abnormal queues to obtain a corresponding first evaluation report. According to the invention, the evaluation efficiency can be improved, and the evaluation real-time performance can be improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a processing method and device for evaluating the perception response performance of a vehicle. Background Art

[0002] When evaluating the perception and response performance of automatic / unmanned vehicles, it is necessary to evaluate the following performance of the vehicle in many aspects, such as the perception and response performance of traffic signs (such as speed limit signs, U-turn signs, entrance and exit signs, stop and yield signs, traffic lights, etc.), the perception and response performance of road scenes (such as curves, ramps, tunnels, toll or parking places with lifting poles, various lanes, various intersections, etc.). The current conventional processing method for evaluating these performances is to first collect and save real-time data during the vehicle's driving process, and then analyze and score the collected data based on different performance evaluation purposes based on manual or semi-manual methods after the vehicle has finished driving. Obviously, this conventional processing method is difficult to effectively improve its evaluation efficiency due to the existence of manual operation processes. Summary of the invention

[0003] The purpose of the present invention is to provide a processing method, device, electronic device and computer-readable storage medium for evaluating the perception response performance of a vehicle in view of the defects of the prior art. When the first vehicle starts the evaluation, the present invention initializes two empty data collection queues as corresponding vehicle data queues and perception data queues, and initializes fourteen empty exception queues to form corresponding fourteen types of exception queues; and during the evaluation of the first vehicle, the real-time motion, positioning and road information output by the vehicle system is collected to generate corresponding vehicle data records and added to the vehicle data queue, and the target perception data output by the vehicle system is collected to generate corresponding perception data records and added to the perception data queue; and during the evaluation of the first vehicle, fourteen types of perception response anomaly recognition processing are performed according to the vehicle data queue, the perception data queue and the fourteen types of exception queues; and after the first vehicle ends the evaluation, the fourteen types of exception queues are comprehensively scored to obtain the corresponding first evaluation report. Through the present invention, not only can the manual operation process be eliminated, but also real-time evaluation can be performed during vehicle driving, thereby achieving the purpose of improving evaluation efficiency and evaluation real-time performance.

[0004] To achieve the above object, a first aspect of an embodiment of the present invention provides a processing method for evaluating the perception response performance of a vehicle, the method comprising:

[0005] When the first vehicle starts the evaluation, two empty data collection queues are initialized and recorded as the corresponding vehicle data queue and perception data queue; and fourteen empty exception queues are initialized to form corresponding fourteen types of exception queues; the fourteen types of exception queues include speed limit exception queue, U-turn exception queue, main and auxiliary road switching exception queue, stop and yield exception queue, first type signal light exception queue, second type signal light exception queue, third type signal light exception queue, curve exception queue, ramp exception queue, tunnel exception queue, lift pole exception queue, lane exception queue, no light intersection exception queue and roundabout exception queue;

[0006] During the evaluation of the first vehicle, data collection is performed on the real-time motion, positioning and road information output by the vehicle system to generate corresponding vehicle data records and add them to the vehicle data queue; and data collection is performed on the target perception data output by the vehicle system to generate corresponding perception data records and add them to the perception data queue;

[0007] During the evaluation of the first vehicle, fourteen types of perception response anomaly identification processing are performed according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues; the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and yield anomalies, type I traffic light anomalies, type II traffic light anomalies, type III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies and roundabout anomalies;

[0008] After the first vehicle is evaluated, a comprehensive score is calculated based on the fourteen types of abnormal queues to obtain a corresponding first evaluation report.

[0009] Preferably, the vehicle data queue is composed of a plurality of vehicle data records; the vehicle data record is composed of a corresponding first timestamp, vehicle speed, acceleration, heading angle, vehicle head center coordinates, rear axle center coordinates, road curvature, road section lane markings, traffic intersection type, and lane speed upper and lower limits; the traffic intersection type includes straight road / meeting intersection;

[0010] The vehicle speed, the acceleration, the heading angle, the center coordinates of the front of the vehicle, the center coordinates of the rear axle, the road curvature, the lane sign, the traffic intersection type and the upper and lower limits of lane speed are the real-time driving speed of the vehicle, the real-time acceleration of the vehicle, the real-time heading angle of the vehicle, the real-time coordinates of the center point of the top of the front of the vehicle, the real-time coordinates of the center point of the rear axle of the vehicle, the curvature information of the corresponding point on the high-precision map of the intersection of the center point of the rear axle of the vehicle and the vertical line of the road surface, the road sign + section sign + lane sign corresponding to the lane where the vehicle is located, the classification information of the traffic intersection where the vehicle is located, and the upper and lower speed limit thresholds of the lane where the vehicle is located.

[0011] Preferably, the perception data queue is composed of a plurality of the perception data records; the perception data record is composed of a corresponding second timestamp and a plurality of perception targets; the perception target is composed of a corresponding target identifier, target type and target attribute;

[0012] The target identifier is a unique identifier assigned to each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module;

[0013] The target type is the identification object type given for each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module, including at least vehicles, obstacles, lifting poles, U-turn signs, road entrance and exit signs, speed limit signs, stop and yield signs, motor vehicle signal lights, direction indicator lights, lane signal lights, stop and yield lines, and stop lines;

[0014] The target attribute is a set of object attributes given to each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module, and is composed of two categories: general attributes and individual attributes; the general attributes are common attributes possessed by each target type; the individual attributes are personalized attributes unique to some target types;

[0015] The general attributes include coordinates, speed, direction, target frame vertex set, target frame size and target road section lane identification; the coordinates, speed, direction, target frame vertex set and target frame size are the position coordinates of the target ground center point, the relative motion speed between the target and the vehicle, the relative motion direction between the target and the vehicle, the vertex coordinate set of the target corresponding detection frame and the geometric size of the target corresponding detection frame given for each target when the vehicle-mounted system detects and identifies the targets around the vehicle body through the built-in perception module; the target road section lane identification is the road identification + section identification + lane identification of the target corresponding lane given when the vehicle-mounted system refreshes the corresponding lane information of each target around the vehicle body through the built-in perception module and the high-precision map module;

[0016] When the target type is a vehicle, an obstacle, a U-turn sign, a stop sign, a stop line or a stop line, the corresponding individual attribute is empty;

[0017] When the target type is a lifting rod, the corresponding individual attribute includes a lifting state; the lifting state includes lifting the rod and lowering the rod;

[0018] When the target type is a road entrance and exit sign, the corresponding individual attribute includes the entrance and exit type; the entrance and exit type includes exit and entrance;

[0019] When the target type is a speed limit sign, the corresponding individual attributes include the start and end speed limit type and the start and end speed limit value; the start and end speed limit type includes start and end; the start and end speed limit value corresponds to the upper speed limit threshold;

[0020] When the target type is a motor vehicle signal light, a direction indicator light or a lane signal light, the corresponding individual attributes include the signal light status; the signal light status corresponding to the motor vehicle signal light and the direction indicator light includes red light and green light; the signal light status corresponding to the lane signal light includes pass and no pass.

[0021] Preferably, when the speed limit exception queue is not empty, it is composed of a plurality of speed limit exception records; each of the speed limit exception records corresponds to a speed limit exception event; the speed limit exception record is composed of a corresponding first moment and a speed limit exception type; the first moment is the time information corresponding to the speed limit exception event; the speed limit exception type is the type information corresponding to the speed limit exception event, including exceeding the speed limit and speed being too low;

[0022] When the U-turn abnormality queue is not empty, it is composed of a plurality of U-turn abnormality records; each of the U-turn abnormality records corresponds to a U-turn abnormality event; the U-turn abnormality record is composed of a corresponding first time period and a U-turn abnormality type; the first time period is the time period information corresponding to the current U-turn abnormality event; the U-turn abnormality type is the type information corresponding to the current U-turn abnormality event, including excessive speed and excessive time;

[0023] When the main-auxiliary road switching abnormality queue is not empty, it is composed of multiple main-auxiliary road switching abnormality records; each of the main-auxiliary road switching abnormality records corresponds to a main-auxiliary road switching abnormality event; the main-auxiliary road switching abnormality record is composed of the corresponding second time period and the main-auxiliary road switching abnormality type; the second time period is the time period information corresponding to the current main-auxiliary road switching abnormality event; the main-auxiliary road switching abnormality type is the type information corresponding to the current main-auxiliary road switching abnormality event, including main-to-auxiliary without deceleration and auxiliary-to-main without acceleration;

[0024] When the stop and yield abnormality queue is not empty, it is composed of a plurality of stop and yield abnormality records; each of the stop and yield abnormality records corresponds to a stop and yield abnormality event; the stop and yield abnormality record is composed of a corresponding third time period and a stop and yield abnormality type; the third time period is the time period information corresponding to the stop and yield abnormality event; the stop and yield abnormality type is the type information corresponding to the stop and yield abnormality event, including stop and yield crossing the line, being too far from the stop and yield line, and parking for too long;

[0025] When the first type of signal light abnormality queue is not empty, it is composed of a plurality of first type of signal light abnormality records; each of the first type of signal light abnormality records corresponds to a motor vehicle signal light abnormality event; the first type of signal light abnormality record is composed of a corresponding fourth time period and a first type of signal light abnormality type; the fourth time period is the time period information corresponding to the motor vehicle signal light abnormality event; the first type of signal light abnormality type is the type information corresponding to the motor vehicle signal light abnormality event, including going straight at a red light, turning left at a red light without stopping, turning right at a red light and stopping, stopping at a red light too far from the stop line, and starting time of a green light too long;

[0026] When the second-category signal light abnormality queue is not empty, it is composed of a plurality of second-category signal light abnormality records; each of the second-category signal light abnormality records corresponds to a direction indicator signal light abnormality event; the second-category signal light abnormality record is composed of the corresponding fifth time period and the second-category signal light abnormality type; the fifth time period is the time period information corresponding to the current direction indicator signal light abnormality event; the second-category signal light abnormality type is the type information corresponding to the current direction indicator signal light abnormality event, including going straight at a red light, turning left at a red light without stopping, turning right at a red light and stopping, stopping at a red light too far from the stop line, and starting time of a green light too long;

[0027] When the three types of signal light abnormality queues are not empty, they are composed of a plurality of three types of signal light abnormality records; each of the three types of signal light abnormality records corresponds to a lane signal light abnormality event; the three types of signal light abnormality records are composed of the corresponding sixth time period and three types of signal light abnormality types; the sixth time period is the time period information corresponding to the lane signal light abnormality event; the three types of signal light abnormality types are the type information corresponding to the lane signal light abnormality event, including no-stop, no-lane-changing and stop;

[0028] When the curve abnormality queue is not empty, it is composed of a plurality of curve abnormality records; each of the curve abnormality records corresponds to a curve abnormality event; the curve abnormality record is composed of the corresponding seventh time period and the curve abnormality type; the seventh time period is the time period information corresponding to the current curve abnormality event; the curve abnormality type is the type information corresponding to the current curve abnormality event, including curve overspeeding, curve underspeed and curve parking;

[0029] When the ramp abnormality queue is not empty, it is composed of a plurality of ramp abnormality records; each ramp abnormality record corresponds to a ramp abnormality event; the ramp abnormality record is composed of the corresponding eighth time period and the ramp abnormality type; the eighth time period is the time period information corresponding to the ramp abnormality event; the ramp abnormality type is the type information corresponding to the ramp abnormality event, including ramp speeding, ramp low speed and ramp parking;

[0030] When the tunnel abnormality queue is not empty, it is composed of a plurality of tunnel abnormality records; each of the tunnel abnormality records corresponds to a tunnel abnormality event; the tunnel abnormality record is composed of a corresponding ninth time period and a tunnel abnormality type; the ninth time period is the time period information corresponding to the tunnel abnormality event; the tunnel abnormality type is the type information corresponding to the tunnel abnormality event, including tunnel lane change and tunnel parking;

[0031] When the lifting rod abnormality queue is not empty, it is composed of a plurality of lifting rod abnormality records; each lifting rod abnormality record corresponds to a lifting rod abnormality event; the lifting rod abnormality record is composed of the corresponding tenth time period and the lifting rod abnormality type; the tenth time period is the time period information corresponding to the lifting rod abnormality event; the lifting rod abnormality type is the type information corresponding to the lifting rod abnormality event, including lowering the rod without stopping and raising the rod stopping;

[0032] When the lane abnormality queue is not empty, it is composed of a plurality of lane abnormality records; each lane abnormality record corresponds to a lane abnormality event; the lane abnormality record is composed of a corresponding eleventh time period and a lane abnormality type; the eleventh time period is the time period information corresponding to the lane abnormality event; the lane abnormality type is the type information corresponding to the lane abnormality event, including passing during non-passing time periods, not bypassing stationary vehicles, not bypassing obstacles, vehicle collision and abnormal parking;

[0033] When the no-light intersection abnormality queue is not empty, it is composed of a plurality of no-light intersection abnormality records; each of the no-light intersection abnormality records corresponds to an no-light intersection abnormality event; the no-light intersection abnormality record is composed of a corresponding twelfth time period and an no-light intersection abnormality type; the twelfth time period is the time period information corresponding to the no-light intersection event; the no-light intersection abnormality type is the type information corresponding to the no-light intersection abnormal event, including right turn without giving way to left straight ahead, left turn without giving way to right straight ahead, and straight ahead without giving way to right left turn or right straight ahead;

[0034] When the roundabout exception queue is not empty, it is composed of multiple roundabout exception records; each of the roundabout exception records corresponds to a roundabout exception event; the roundabout exception record is composed of a corresponding thirteenth time period and a roundabout exception type; the thirteenth time period is the time period information corresponding to the roundabout event; the roundabout exception type is the type information corresponding to the roundabout exception event, including failure to exit according to the planned exit, emergency braking or parking when entering the intersection, and emergency braking or parking when exiting the intersection.

[0035] Preferably, the performing fourteen types of perception response anomaly identification processing according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues specifically includes:

[0036] Step 501, identifying whether the vehicle has two abnormal conditions, namely, driving too fast or driving too slow, in a known lane speed limit range according to the vehicle data queue and the perception data queue, and updating the corresponding speed limit abnormality queue based on the identification result;

[0037] Step 502, identifying whether the vehicle has two abnormal conditions, namely, a too fast turning speed and a too long turning time, during a U-turn process according to the vehicle data queue and the perception data queue, and updating the corresponding U-turn abnormality queue based on the identification result;

[0038] Step 503, identifying whether the vehicle has two abnormal conditions, namely, not decelerating when entering the auxiliary road and not accelerating when entering the main road, during the main-auxiliary road switching process according to the vehicle data queue and the perception data queue, and updating the corresponding main-auxiliary road switching abnormality queue based on the identification result;

[0039] Step 504, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle has any of the following three abnormal conditions during the stop and yield process: failure to stop in time resulting in crossing the line, premature stopping resulting in the parking position being too far from the stop and yield line, and parking time being too long, and updating the corresponding stop and yield abnormality queue based on the identification results;

[0040] Step 505, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight at a known red light, continues to turn left at a known red light, stops at a wrong right turn at a red light, stops too early at a red light resulting in the parking position being too far from the stop line, and takes too long to start the vehicle at a green light when the vehicle signal light status is known, and updating the corresponding signal light abnormality queue according to the identification result;

[0041] Step 506, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight when the direction indicator light is known, continues to turn left when the direction indicator light is known, stops when turning right at a red light, stops too early at a red light so that the parking position is too far from the stop line, and takes too long to start the vehicle when the direction indicator light is known, and updating the corresponding two types of signal light abnormality queues based on the identification results;

[0042] Step 507, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight when the lane is known to be prohibited, stops without changing lanes when the lane is prohibited, and stops when the lane is known to be passable when the vehicle is stopped when the lane is known to be passable under three abnormal conditions, and updating the corresponding three types of abnormal signal light queues based on the identification results;

[0043] Step 508, identifying whether the vehicle has three abnormal conditions, namely, excessive speed, low speed, and parking on a curve, when driving on a curve according to the vehicle data queue, and updating the corresponding curve abnormality queue based on the identification result;

[0044] Step 509, identifying whether the vehicle is driving on the ramp at an excessively high speed, driving at an excessively low speed, or parking on the ramp according to the vehicle data queue, and updating the corresponding ramp abnormality queue based on the identification result;

[0045] Step 510, identifying whether two abnormal conditions, namely, lane changing and parking in the tunnel, occur when the vehicle is traveling in the tunnel according to the vehicle data queue, and updating the corresponding tunnel abnormality queue based on the identification result;

[0046] Step 511, identifying whether the vehicle has two abnormal conditions, namely, lowering the bar without deceleration and raising the bar without acceleration, when passing through a toll station entrance or a parking lot entrance with a lifting bar, based on the vehicle data queue and the perception data queue, and updating the corresponding lifting bar abnormality queue based on the identification result;

[0047] Step 512, identifying five abnormal conditions when the vehicle is traveling on the lane, including occupying the lane during non-traffic hours, not detouring around stationary vehicles, not detouring around obstacles, colliding with other moving vehicles around the vehicle, and parking on the lane, according to the vehicle data queue and the perception data queue, and updating the corresponding lane abnormality queue based on the identification results;

[0048] Step 513, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle fails to yield to a left straight vehicle during a right turn, fails to yield to a right straight vehicle during a left turn, or fails to yield to a right left-turning or straight vehicle during a straight drive when the vehicle is at a no-light intersection, and updating the corresponding no-light intersection abnormality queue based on the identification results;

[0049] Step 514, based on the vehicle data queue and the perception data queue, identify three types of abnormal conditions when the vehicle is at a roundabout: the vehicle fails to turn out according to the planned exit, the vehicle emergency brakes / stops when entering the intersection, and the vehicle emergency brakes / stops when turning out of the intersection, and updates the corresponding roundabout abnormal queue based on the identification results.

[0050] Preferably, the first evaluation report corresponding to the fourteen types of abnormal queues is obtained by comprehensive scoring, specifically including:

[0051] The number of each type of abnormality of the fourteen types of abnormal queues is counted to obtain the corresponding number of fourteen types of abnormalities; the fourteen types of abnormalities include the number of speed limit abnormalities, the number of U-turn abnormalities, the number of main and auxiliary road switching abnormalities, the number of stop and yield abnormalities, the number of type I traffic lights abnormalities, the number of type II traffic lights abnormalities, the number of type III traffic lights abnormalities, the number of curve abnormalities, the number of ramp abnormalities, the number of tunnel abnormalities, the number of lifting pole abnormalities, the number of lane abnormalities, the number of unlit intersection abnormalities and the number of roundabout abnormalities;

[0052] And the abnormal type corresponding to each abnormal number that is not 0 in the fourteen types of abnormal numbers is recorded as a corresponding abnormal evaluation type;

[0053] And based on the preset first correspondence table, the abnormal deduction score corresponding to each abnormal evaluation type is identified to obtain the corresponding first deduction value; the first correspondence table is a correspondence table that reflects the correspondence between the abnormal type and the abnormal deduction value; the first correspondence table includes fourteen first correspondence records; the first correspondence record includes a first abnormal type field and a first deduction value field; the first abnormal type field includes the speed limit abnormal type, the U-turn abnormal type, the main and auxiliary road switching abnormal type, the stop and yield abnormal type, the first type of signal light abnormal type, the second type of signal light abnormal type, the third type of signal light abnormal type, the curve abnormal type, the ramp abnormal type, the tunnel abnormal type, the lifting pole abnormal type, the lane abnormal type, the unlighted intersection abnormal type and the roundabout abnormal type; the first deduction value field is an integer score;

[0054] The sum of all the first deduction values ​​obtained is calculated to obtain a corresponding total deduction value; and the corresponding evaluation score for the time being is calculated based on the preset full score and the total deduction value = the full score - the total deduction value; and a corresponding first evaluation record is formed by each of the abnormal evaluation types and the corresponding first deduction values; and the corresponding first evaluation report is formed by all the first evaluation records obtained and the current evaluation score.

[0055] A second aspect of an embodiment of the present invention provides a device for implementing the processing method for evaluating the perception response performance of a vehicle as described in the first aspect above, wherein the device is connected to an onboard system of a first vehicle; the device comprises: a data acquisition module, a data storage module, an abnormality identification module and an abnormality evaluation module;

[0056] The data acquisition module is used to initialize two empty data acquisition queues as corresponding vehicle data queues and perception data queues and store them in the data storage module when the first vehicle starts the evaluation; and initialize fourteen empty exception queues to form corresponding fourteen types of exception queues and store them in the data storage module; the fourteen types of exception queues include speed limit exception queue, U-turn exception queue, main and auxiliary road switching exception queue, stop and yield exception queue, first type signal light exception queue, second type signal light exception queue, third type signal light exception queue, curve exception queue, ramp exception queue, tunnel exception queue, lifting pole exception queue, lane exception queue, no light intersection exception queue and roundabout exception queue;

[0057] The data acquisition module is also used to collect data on the real-time motion, positioning and road information output by the vehicle-mounted system during the evaluation of the first vehicle, generate corresponding vehicle data records and add them to the vehicle data queue; and collect data on the target perception data output by the vehicle-mounted system to generate corresponding perception data records and add them to the perception data queue;

[0058] The anomaly recognition module is used to perform fourteen types of perception response anomaly recognition processing according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues during the evaluation process of the first vehicle; the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and yield anomalies, type I traffic light anomalies, type II traffic light anomalies, type III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies and roundabout anomalies;

[0059] The abnormal evaluation module is used to obtain a corresponding first evaluation report by performing comprehensive scoring according to the fourteen types of abnormal queues after the first vehicle completes the evaluation.

[0060] A third aspect of an embodiment of the present invention provides an electronic device, including: a memory, a processor, and a transceiver;

[0061] The processor is used to be coupled to the memory, read and execute instructions in the memory, so as to implement the method steps described in the first aspect above;

[0062] The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

[0063] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions. When the computer instructions are executed by a computer, the computer executes the instructions of the method described in the first aspect above.

[0064] The embodiment of the present invention provides a processing method, device, electronic device and computer-readable storage medium for evaluating the perception response performance of a vehicle. It can be seen from the above invention content that when the first vehicle starts the evaluation, the embodiment of the present invention initializes two empty data collection queues as corresponding vehicle data queues and perception data queues, and initializes fourteen empty exception queues to form corresponding fourteen types of exception queues; and during the evaluation of the first vehicle, the real-time motion, positioning and road information output by the vehicle system is collected to generate corresponding vehicle data records and added to the vehicle data queue, and the target perception data output by the vehicle system is collected to generate corresponding perception data records and added to the perception data queue; and during the evaluation of the first vehicle, fourteen types of perception response anomaly recognition processing are performed according to the vehicle data queue, perception data queue and fourteen types of exception queues; and after the first vehicle ends the evaluation, the fourteen types of exception queues are comprehensively scored to obtain the corresponding first evaluation report. The embodiment of the present invention not only gets rid of the manual operation process, but also realizes real-time evaluation during vehicle driving, thereby improving the evaluation efficiency and the real-time performance of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of a processing method for evaluating the perception response performance of a vehicle provided in the first embodiment of the present invention;

[0066] Figure 2 A schematic diagram of the data structure of a vehicle data queue and a perception data queue according to the first embodiment of the present invention;

[0067] Figure 3 This is a schematic diagram of the data structure of fourteen types of abnormal queues according to the first embodiment of the present invention;

[0068] Figure 4 A module structure diagram of a processing device for evaluating the perception response performance of a vehicle provided in the second embodiment of the present invention;

[0069] Figure 5 A schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] A processing method for evaluating the perception response performance of a vehicle is provided in a first embodiment of the present invention, such as Figure 1 A schematic diagram of a processing method for evaluating the perception response performance of a vehicle provided in the first embodiment of the present invention is shown, and includes the following main steps:

[0072] Step 1, when the first vehicle starts the evaluation, initialize two empty collection data queues to be recorded as the corresponding vehicle data queue and perception data queue; and initialize fourteen empty exception queues to form corresponding fourteen types of exception queues.

[0073] Here, the first vehicle in the embodiment of the present invention is an automatic / unmanned driving vehicle to be evaluated.

[0074] Step 2. During the evaluation of the first vehicle, data collection is performed on the real-time motion, positioning and road information output by the vehicle-mounted system to generate corresponding vehicle data records, which are added to the vehicle data queue; and data collection is performed on the target perception data output by the vehicle-mounted system to generate corresponding perception data records, which are added to the perception data queue.

[0075] Here, the data collection queues of the embodiment of the present invention have two: vehicle data queue and perception data queue. The real-time evaluation queues of the embodiment of the present invention have fourteen categories, namely fourteen abnormal queues: speed limit abnormal queue, U-turn abnormal queue, main and auxiliary road switching abnormal queue, stop and yield abnormal queue, type I signal light abnormal queue, type II signal light abnormal queue, type III signal light abnormal queue, curve abnormal queue, ramp abnormal queue, tunnel abnormal queue, lifting pole abnormal queue, lane abnormal queue, no light intersection abnormal queue and roundabout abnormal queue.

[0076] like Figure 2 As shown in the data structure diagram of the vehicle data queue and the perception data queue of the first embodiment of the present invention, the vehicle data queue of the embodiment of the present invention is composed of multiple vehicle data records; each vehicle data record is composed of a corresponding first timestamp, vehicle speed, acceleration, heading angle, vehicle head center coordinates, rear axle center coordinates, road curvature, road section lane markings, traffic intersection type and lane speed upper and lower limits; traffic intersection types include straight roads / meeting intersections; among them, vehicle speed, acceleration, heading angle, vehicle head center coordinates, rear axle center coordinates, road curvature, lane markings, traffic intersection type and vehicle The upper and lower limits of lane speed are the real-time speed of the vehicle, the real-time acceleration of the vehicle, the real-time heading angle of the vehicle, the real-time coordinates of the center point of the top of the vehicle's front, the real-time coordinates of the center point of the rear axle of the vehicle, the curvature information of the intersection of the center point of the rear axle of the vehicle and the vertical line of the road surface at the corresponding point on the high-precision map, the road sign + section sign + lane sign corresponding to the lane where the vehicle is located, the classification information of the traffic intersection where the vehicle is located, and the upper and lower speed limit thresholds of the lane where the vehicle is located.

[0077] like Figure 2 As shown, the perception data queue of the embodiment of the present invention is composed of multiple perception data records; each perception data record is composed of a corresponding second timestamp and multiple perception targets; each perception target is composed of a corresponding target identifier, target type and target attribute; wherein:

[0078] 1) The target identifier is a unique identifier assigned to each target when the vehicle system identifies the targets around the vehicle body through the built-in perception module;

[0079] 2) Target type refers to the type of identification object given by the vehicle system for each target when it identifies the targets around the vehicle body through the built-in perception module, including at least vehicles, obstacles, lifting poles, U-turn signs, road entrance and exit signs, speed limit signs, stop and yield signs, motor vehicle signal lights, direction indicator lights, lane signal lights, stop and yield lines, and stop lines;

[0080] 3) Target attributes are the set of object attributes given to each target by the vehicle system when it uses the built-in perception module to identify the targets around the vehicle body. They are composed of two categories: general attributes and individual attributes. General attributes are common attributes that each target type has; individual attributes are personalized attributes that are unique to some target types. Specifically:

[0081] a. Common attributes include coordinates, speed, orientation, target frame vertex set, target frame size, and target road section lane identification; among them, coordinates, speed, orientation, target frame vertex set, and target frame size are the position coordinates of the target ground center point, the relative motion speed between the target and the vehicle, the relative motion direction between the target and the vehicle, the vertex coordinate set of the target corresponding detection frame, and the geometric size of the target corresponding detection frame given for each target when the vehicle-mounted system detects and identifies the targets around the vehicle body through the built-in perception module; the target road section lane identification is the road identification + section identification + lane identification of the target corresponding lane given when the vehicle-mounted system refreshes the corresponding lane information of each target around the vehicle body through the built-in perception module and the high-precision map module;

[0082] b. When the target type is a vehicle, obstacle, U-turn sign, stop and yield sign, stop and yield line or stop line, the corresponding individual attributes are empty; when the target type is a lifting pole, the corresponding individual attributes include the lifting state, which includes raising the pole and lowering the pole; when the target type is a road entrance and exit sign, the corresponding individual attributes include the entrance and exit type, which includes exit and entrance; when the target type is a speed limit sign, the corresponding individual attributes include the start and end speed limit type and the start and end speed limit value, the start and end speed limit type includes start and end, and the start and end speed limit value is the corresponding upper speed threshold; when the target type is a motor vehicle signal light, a direction indicator signal light or a lane signal light, the corresponding individual attributes all include the signal light state. It should be noted that the signal light states corresponding to the motor vehicle signal light and the direction indicator signal light include red light and green light, and the signal light states corresponding to the lane signal light include pass and no pass.

[0083] like Figure 3 As shown in the data structure diagram of the fourteen types of abnormal queues of the first embodiment of the present invention, the data structure of the fourteen types of abnormal queues is described as follows:

[0084] 1) When the speed limit exception queue is not empty, it is composed of multiple speed limit exception records; each speed limit exception record corresponds to a speed limit exception event; the speed limit exception record is composed of the corresponding first moment and the speed limit exception type; the first moment is the time information corresponding to the speed limit exception event; the speed limit exception type is the type information corresponding to the speed limit exception event, including exceeding the speed limit and speed too low;

[0085] 2) When the U-turn abnormality queue is not empty, it is composed of multiple U-turn abnormality records; each U-turn abnormality record corresponds to a U-turn abnormal event; the U-turn abnormality record is composed of the corresponding first time period and the U-turn abnormality type; the first time period is the time period information corresponding to the current U-turn abnormal event; the U-turn abnormality type is the type information corresponding to the current U-turn abnormal event, including excessive speed and too long time;

[0086] 3) When the main-auxiliary road switching abnormality queue is not empty, it is composed of multiple main-auxiliary road switching abnormality records; each main-auxiliary road switching abnormality record corresponds to a main-auxiliary road switching abnormality event; the main-auxiliary road switching abnormality record is composed of the corresponding second time period and the main-auxiliary road switching abnormality type; the second time period is the time period information corresponding to the current main-auxiliary road switching abnormality event; the main-auxiliary road switching abnormality type is the type information corresponding to the current main-auxiliary road switching abnormality event, including main-auxiliary road output without deceleration and auxiliary-main road input without acceleration;

[0087] 4) When the stop and yield abnormality queue is not empty, it is composed of multiple stop and yield abnormality records; each stop and yield abnormality record corresponds to a stop and yield abnormality event; the stop and yield abnormality record is composed of the corresponding third time period and the stop and yield abnormality type; the third time period is the time period information corresponding to the stop and yield abnormality event; the stop and yield abnormality type is the type information corresponding to the stop and yield abnormality event, including stop and yield crossing the line, too far from the stop and yield line, and too long parking time;

[0088] 5) When the first-class signal light abnormality queue is not empty, it is composed of multiple first-class signal light abnormality records; each first-class signal light abnormality record corresponds to a motor vehicle signal light abnormality event; the first-class signal light abnormality record is composed of the corresponding fourth time period and the first-class signal light abnormality type; the fourth time period is the time period information corresponding to the motor vehicle signal light abnormality event; the first-class signal light abnormality type is the type information corresponding to the motor vehicle signal light abnormality event, including red light straight ahead, red light left turn without stopping, red light right turn stop, red light stop too far from the stop line and green light start time too long;

[0089] 6) When the Class II signal light abnormality queue is not empty, it is composed of multiple Class II signal light abnormality records; each Class II signal light abnormality record corresponds to a direction indicator signal light abnormality event; the Class II signal light abnormality record is composed of the corresponding fifth time period and the Class II signal light abnormality type; the fifth time period is the time period information corresponding to the current direction indicator signal light abnormality event; the Class II signal light abnormality type is the type information corresponding to the current direction indicator signal light abnormality event, including red light straight ahead, red light left turn without stopping, red light right turn stop, red light stop too far from the stop line and green light start time too long;

[0090] 7) When the three-type signal light abnormality queue is not empty, it is composed of multiple three-type signal light abnormality records; each three-type signal light abnormality record corresponds to a lane signal light abnormality event; the three-type signal light abnormality record is composed of the corresponding sixth time period and three-type signal light abnormality types; the sixth time period is the time period information corresponding to the lane signal light abnormality event; the three-type signal light abnormality type is the type information corresponding to the lane signal light abnormality event, including no-stop, no-lane-changing and stop;

[0091] 8) When the curve abnormality queue is not empty, it is composed of multiple curve abnormality records; each curve abnormality record corresponds to a curve abnormality event; the curve abnormality record is composed of the corresponding seventh time period and the curve abnormality type; the seventh time period is the time period information corresponding to the current curve abnormality event; the curve abnormality type is the type information corresponding to the current curve abnormality event, including curve overspeeding, curve underspeed and curve parking;

[0092] 9) When the ramp abnormality queue is not empty, it is composed of multiple ramp abnormality records; each ramp abnormality record corresponds to a ramp abnormality event; the ramp abnormality record is composed of the corresponding eighth time period and ramp abnormality type; the eighth time period is the time period information corresponding to the ramp abnormality event; the ramp abnormality type is the type information corresponding to the ramp abnormality event, including ramp speeding, ramp low speed and ramp parking;

[0093] 10) When the tunnel abnormality queue is not empty, it is composed of multiple tunnel abnormality records; each tunnel abnormality record corresponds to a tunnel abnormality event; the tunnel abnormality record is composed of the corresponding ninth time period and tunnel abnormality type; the ninth time period is the time period information corresponding to the tunnel abnormality event; the tunnel abnormality type is the type information corresponding to the tunnel abnormality event, including tunnel lane change and tunnel parking;

[0094] 11) When the lift rod abnormality queue is not empty, it is composed of multiple lift rod abnormality records; each lift rod abnormality record corresponds to a lift rod abnormality event; the lift rod abnormality record is composed of the corresponding tenth time period and lift rod abnormality type; the tenth time period is the time period information corresponding to the lift rod abnormality event; the lift rod abnormality type is the type information corresponding to the lift rod abnormality event, including lowering the rod without stopping and raising the rod stopping;

[0095] 12) When the lane abnormality queue is not empty, it is composed of multiple lane abnormality records; each lane abnormality record corresponds to a lane abnormality event; the lane abnormality record is composed of the corresponding eleventh time period and lane abnormality type; the eleventh time period is the time period information corresponding to the lane abnormality event; the lane abnormality type is the type information corresponding to the lane abnormality event, including passing during non-passing time periods, not bypassing stationary vehicles, not bypassing obstacles, vehicle collision and abnormal parking;

[0096] 13) When the no-light intersection abnormality queue is not empty, it is composed of multiple no-light intersection abnormality records; each no-light intersection abnormality record corresponds to an no-light intersection abnormality event; the no-light intersection abnormality record is composed of the corresponding twelfth time period and the no-light intersection abnormality type; the twelfth time period is the time period information corresponding to the no-light intersection event; the no-light intersection abnormality type is the type information corresponding to the no-light intersection abnormal event, including right turn without giving way to the left straight, left turn without giving way to the right straight and straight driving without giving way to the right left turn or right straight driving;

[0097] 14) When the roundabout exception queue is not empty, it is composed of multiple roundabout exception records; each roundabout exception record corresponds to a roundabout exception event; the roundabout exception record is composed of the corresponding thirteenth time period and the roundabout exception type; the thirteenth time period is the time period information corresponding to the roundabout event; the roundabout exception type is the type information corresponding to the roundabout exception event, including failure to exit according to the planned exit, emergency braking or parking when entering the intersection, and emergency braking or parking when exiting the intersection.

[0098] Step 3, during the evaluation process of the first vehicle, fourteen types of perception response anomaly identification processing are performed according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues;

[0099] Among them, the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and yield anomalies, type I traffic light anomalies, type II traffic light anomalies, type III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies, and roundabout anomalies;

[0100] Specifically, step 3-1, identifying whether the vehicle is driving too fast or too slow in a known lane speed limit range according to the vehicle data queue and the perception data queue, and updating the corresponding speed limit abnormality queue based on the identification result;

[0101] Specifically, it includes: step 3-1-1, when a vehicle data record is added to the vehicle data queue, the vehicle data record added at that time is recorded as the corresponding current vehicle record; and the perception data record in the perception data queue that is closest to the current vehicle record time is recorded as the corresponding current perception record;

[0102] Step 3-1-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a speed limit sign to obtain a corresponding current confirmation result;

[0103] Among them, the current confirmation results include existence and non-existence;

[0104] Step 3-1-3, identify the current confirmation result; if the current confirmation result is yes, go to step 3-1-4; if the current confirmation result is no, go to step 3-1-5;

[0105] Step 3-1-4, match the target road section lane marking in the current perception record with the road section lane marking recorded by the current vehicle and record the perception target whose target type is a speed limit sign as the corresponding current speed limit sign; and identify the start and end speed limit types of the current speed limit sign; if the current start and end speed limit types are end, go to step 3-1-5; if the current start and end speed limit types are start, extract the coordinates of the current speed limit sign and the start and end speed limit values ​​as the corresponding current sign coordinates and current sign speed limit thresholds, and confirm whether the center point of the top of the vehicle's front has entered the speed limit section of the current speed limit sign based on the center coordinates of the front of the current vehicle record and the current sign coordinates. If it is confirmed that the vehicle has not entered, go to step 3-1-5. If it is confirmed that the vehicle has entered, confirm whether the speed recorded by the current vehicle has exceeded the current speed limit threshold. If it is confirmed that it has exceeded, set the corresponding speed limit exception type to exceed the speed limit and go to step 3-1-6. If it is confirmed that it has not exceeded, calculate the ratio of the current vehicle speed to the current speed limit threshold and identify whether the obtained ratio is lower than the preset first ratio threshold. If the current ratio is lower than the first ratio threshold, set the corresponding speed limit exception type to too low speed and go to step 3-1-6. If the current ratio is not lower than the first ratio threshold, set the corresponding speed limit exception type to no exception and go to step 3-1-6.

[0106] Here, the first ratio threshold is a preset ratio;

[0107] Step 3-1-5, record the speed recorded by the current vehicle and the upper and lower speed thresholds of the lane speed as the corresponding current speed, current upper threshold and current lower threshold; and compare the current speed based on the current upper threshold and current lower threshold; if the current speed is lower than the current lower threshold, set the corresponding speed limit abnormality type to too low speed; if the current speed exceeds the current upper threshold, set the corresponding speed limit abnormality type to exceed the speed limit; if the current speed is between the current lower threshold and the current upper threshold, calculate the difference between the current speed and the current lower threshold and the difference between the current upper threshold and the current lower threshold to obtain the corresponding first difference and second difference, calculate the ratio of the first difference to the second difference and identify whether the obtained ratio is lower than the first ratio threshold, if the current ratio is lower than the first ratio threshold, set the corresponding speed limit abnormality type to too low speed, if the current ratio is not lower than the first ratio threshold, set the corresponding speed limit abnormality type to no abnormality;

[0108] Step 3-1-6, identify the speed limit exception type obtained; if the speed limit exception type is not no exception, use the first timestamp of the current vehicle record as a corresponding first moment; and form a corresponding speed limit exception record composed of the first moment obtained this time and the speed limit exception type and add it to the speed limit exception queue;

[0109] Step 3-2, identifying whether the vehicle has two abnormal conditions during the U-turn process, namely, a U-turn speed that is too fast or a U-turn time that is too long, based on the vehicle data queue and the perception data queue, and updating the corresponding U-turn abnormality queue based on the identification result;

[0110] Specifically, it includes: step 3-2-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record is recorded as the corresponding current vehicle record;

[0111] Step 3-2-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a U-turn sign to obtain a corresponding current confirmation result;

[0112] Among them, the current confirmation results include existence and non-existence;

[0113] Step 3-2-3, identify the current confirmation result; if the current confirmation result is yes, confirm whether the U-turn process of the first vehicle is being tracked; if it is confirmed that the tracking is not started, start tracking the current U-turn process of the first vehicle and initialize an empty sequence as the corresponding U-turn tracking sequence and go to step 3-2-4; if it is confirmed that the tracking is being carried out, go to step 3-2-4; if the current confirmation result is no, set the corresponding U-turn exception type to no exception and go to step 3-2-5;

[0114] Step 3-2-4, add the current vehicle record to the U-turn tracking sequence; and confirm whether the road section lane mark of the current vehicle record has entered the non-U-turn section of the current lane; if it is confirmed that it has not entered the non-U-turn section of the current lane, keep tracking and return to step 3-2-1; if it is confirmed that it has entered the non-U-turn section of the current lane, stop this tracking and identify the maximum, minimum and average vehicle speeds in the U-turn tracking sequence to obtain the corresponding maximum speed, minimum speed and average speed, and when the maximum speed exceeds the upper speed threshold of the corresponding lane speed upper and lower limits, set the corresponding U-turn abnormality type to excessive speed, and When the secondary minimum speed is lower than the lower speed threshold of the corresponding lane speed upper and lower limits or when the secondary average speed is lower than the preset U-turn average speed threshold, the corresponding U-turn abnormality type is set to too long time, and when the secondary maximum speed and the secondary minimum speed both meet the speed range specified by the corresponding lane speed upper and lower limits and when the secondary average speed is not lower than the U-turn average speed threshold, the corresponding U-turn abnormality type is set to no abnormality, and after completing the U-turn abnormality type setting, the first timestamps of the first and last vehicle data records of the U-turn tracking sequence are extracted to form a corresponding first time period, and after obtaining the first time period, the U-turn tracking sequence is deleted and the process goes to step 3-2-5;

[0115] Here, the U-turn average speed threshold is a preset speed value;

[0116] Step 3-2-5, identifying the obtained U-turn abnormal type; if the U-turn abnormal type is not no abnormality, a corresponding U-turn abnormal record is formed by the first time period and the U-turn abnormal type obtained this time and added to the U-turn abnormal queue;

[0117] Step 3-3, identifying whether the vehicle has two abnormal conditions during the main-auxiliary road switching process: not decelerating when entering the auxiliary road or not accelerating when entering the main road, based on the vehicle data queue and the perception data queue, and updating the corresponding main-auxiliary road switching abnormality queue based on the identification result;

[0118] Specifically, it includes: step 3-3-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record is recorded as the corresponding current vehicle record;

[0119] Step 3-3-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a road entrance and exit sign to obtain the corresponding current confirmation result;

[0120] Among them, the current confirmation results include existence and non-existence;

[0121] Step 3-3-3, identify the current confirmation result; if the current confirmation result is existence, go to step 3-3-4; if the current confirmation result is non-existence, set the corresponding main and auxiliary road switching abnormality type to no abnormality and go to step 3-3-7;

[0122] Step 3-3-4, matching the lane sign of the target road section in the current perception record with the lane sign of the road section in the current vehicle record and the perception target whose target type is a road entrance and exit sign as the corresponding current entrance and exit sign; and taking the entrance and exit type of the current entrance and exit sign as the corresponding current entrance and exit type; and identifying the current entrance and exit type; if the current entrance and exit type is an exit, go to step 3-3-5; if the current entrance and exit type is an entrance, go to step 3-3-6;

[0123] Step 3-3-5, based on the preset high-precision road map, query the main road exit lane line and the auxiliary road entrance lane speed limit corresponding to the current entrance and exit sign to obtain the corresponding current exit lane line and current speed limit; and confirm whether the center point of the top of the vehicle head has crossed the current exit lane line based on the center coordinates of the vehicle head recorded by the current vehicle; if it is confirmed that it has not crossed, set the corresponding main and auxiliary road switching abnormality type to no abnormality; if it is confirmed that it has crossed, identify whether the vehicle speed recorded by the current vehicle exceeds the current speed limit, if it has not exceeded, set the corresponding main and auxiliary road switching abnormality type to no abnormality, if it has exceeded, set the corresponding main and auxiliary road switching abnormality type to main exit and auxiliary without deceleration; and go to step 3-3-7;

[0124] Step 3-3-6, based on the high-precision road map, query the auxiliary road entrance lane line and the main road entrance lane speed limit corresponding to the current entrance and exit sign to obtain the corresponding current entrance lane line and current speed limit; and confirm whether the center point of the top of the vehicle's front has crossed the current entrance lane line based on the center coordinates of the current vehicle's front; if it is confirmed that it has not crossed, set the corresponding main-auxiliary road switching abnormality type to no abnormality; if it is confirmed that it has crossed, identify whether the speed recorded by the current vehicle does not exceed the current speed limit, if it has exceeded, set the corresponding main-auxiliary road switching abnormality type to no abnormality, if it has not exceeded, set the corresponding main-auxiliary road switching abnormality type to auxiliary to main without acceleration;

[0125] Step 3-3-7, identify the main-auxiliary road switching abnormality type obtained; if the main-auxiliary road switching abnormality type is not no abnormality, the first timestamp recorded by the current vehicle is used as the corresponding end time of the current time, and the time obtained by subtracting a preset first time length from the end time of the current time is used as the corresponding start time of the current time, and the current start time and the current end time form a corresponding second time period, and the second time period obtained this time and the main-auxiliary road switching abnormality type form a corresponding main-auxiliary road switching abnormality record and add it to the main-auxiliary road switching abnormality queue;

[0126] Here, the first duration is a preset time length;

[0127] Step 3-4, based on the vehicle data queue and the perception data queue, identify whether the vehicle has any of the following three abnormal conditions during the stop and yield process: failure to stop in time resulting in crossing the line, premature stopping resulting in the parking position being too far from the stop and yield line, and parking time being too long, and update the corresponding stop and yield abnormality queue based on the identification results;

[0128] Specifically, it includes: step 3-4-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record is recorded as the corresponding current vehicle record;

[0129] Step 3-4-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a perception target of a stop and yield sign to obtain a corresponding current confirmation result;

[0130] Among them, the current confirmation results include existence and non-existence;

[0131] Step 3-4-3, identify the current confirmation result; if the current confirmation result is existence, the target road section lane sign in the current perception record is matched with the road section lane sign in the current vehicle record and the perception target whose target type is a stop and give way sign is recorded as the corresponding current stop and give way sign, and the perception target whose target type is a stop and give way line corresponding to the current stop and give way sign in the current perception record is recorded as the corresponding current stop and give way line, and go to step 3-4-4; if the current confirmation result is non-existent, set the corresponding stop and give way exception type to no exception and go to step 3-4-5;

[0132] Step 3-4-4, based on the center coordinates of the front of the current vehicle recorded, confirm whether the center point of the top of the front of the own vehicle has crossed the current stop and yield line; if it is confirmed that the center point of the top of the front of the own vehicle has crossed the current stop and yield line, set the corresponding stop and yield exception type to stop and yield crossing, and go to step 3-4-5; if it is confirmed that the center point of the top of the front of the own vehicle has not crossed the current stop and yield line, then confirm whether the speed recorded by the current vehicle exceeds the preset stop speed threshold; if it is confirmed that the current speed exceeds the stop speed threshold, set the corresponding stop and yield exception type to no exception, and go to step 3-4-5; if it is confirmed that the current speed does not exceed the stop speed threshold, then estimate the distance between the center point of the top of the front of the own vehicle and the current stop and yield line based on the center coordinates of the front of the current vehicle recorded, and check whether the estimated result exceeds the preset the first spacing threshold of the parking line; if the estimation result exceeds the first spacing threshold, the corresponding parking and yielding abnormality type is set as too far from the parking and yielding line, and go to step 3-4-5; if the estimation result does not exceed the first spacing threshold, further search the vehicle data queue from the current vehicle record forward one by one and use the first vehicle data record whose speed is not less than the parking speed threshold as the corresponding pre-parking record, and calculate the time interval from the pre-parking record to the current vehicle record to obtain the corresponding current parking time, and identify whether the current parking time exceeds the preset parking and yielding time threshold, if the current parking time exceeds the parking and yielding time threshold, the corresponding parking and yielding abnormality type is set as too long parking time, if the current parking time does not exceed the parking and yielding time threshold, the corresponding parking and yielding abnormality type is set as no abnormality;

[0133] Here, the parking speed threshold is a preset speed value close to 0, the first spacing threshold is a preset distance length, and the parking time threshold is a preset time length;

[0134] Step 3-4-5, identifying the obtained parking and yielding exception type; if the parking and yielding exception type is not no exception, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset second time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding third time period by the start time and the end time of the current time, and forming a corresponding parking and yielding exception record by the third time period obtained this time and the parking and yielding exception type and adding it to the parking and yielding exception queue;

[0135] Here, the second duration is a preset time length;

[0136] Step 3-5, based on the vehicle data queue and the perception data queue, identify whether the vehicle continues to go straight at a known red light, continues to turn left at a known red light, stops at a wrong right turn at a red light, stops too early at a red light resulting in the parking position being too far from the stop line, and takes too long to start the vehicle at a green light when the vehicle signal light status is known, and update the corresponding signal light abnormality queue based on the identification results;

[0137] Specifically, it includes: step 3-5-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record is recorded as the corresponding current vehicle record;

[0138] Step 3-5-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a perception target of a motor vehicle signal light to obtain a corresponding current confirmation result;

[0139] Among them, the current confirmation results include existence and non-existence;

[0140] Step 3-5-3, identify the current confirmation result; if the current confirmation result is existence, the target road section lane marking in the current perception record matches the road section lane marking recorded by the current vehicle and the perception target of the target type is a motor vehicle signal light is recorded as the corresponding current signal light, and the perception target of the target type in the current perception record is a stop line and the closest to the rear axle center coordinates of the current vehicle record is recorded as the corresponding current stop line, and the signal light state of the current signal light is recorded as the corresponding current signal light state, and go to step 3-5-4; if the current confirmation result is non-existent, set the corresponding type of signal light abnormality type to no abnormality and go to step 3-5-7;

[0141] Step 3-5-4, taking the speed recorded by the current vehicle as the corresponding current speed; and confirming whether the road section corresponding to the lane sign of the road section recorded by the current vehicle is a left turn, a right turn or a straight section by querying the high-precision road map to obtain the corresponding current section driving direction; and identifying whether the center coordinates of the vehicle head recorded by the current vehicle have crossed the current stop line, if the center coordinates of the vehicle head have crossed the current stop line, setting the corresponding current crossing state to crossed, if the center coordinates of the vehicle head have not crossed the current stop line, setting the corresponding current crossing state to not crossed, and estimating the shortest distance between the center coordinates of the current vehicle head and the current stop line to obtain the corresponding current distance;

[0142] The current driving direction of the road section includes left turn, right turn and straight ahead; the current crossing status includes crossing the line and not crossing the line;

[0143] Step 3-5-5, when the current traffic light state is red, if the current road section driving direction is left turn or straight ahead, and the current crossing line state is not crossing the line, and the current vehicle speed does not exceed the parking speed threshold, and the current spacing exceeds the preset second spacing threshold, then the corresponding type of traffic light abnormality is set to red light stop too far from the stop line; if the current road section driving direction is left turn or straight ahead, and the current crossing line state is not crossing the line, and the current vehicle speed does not exceed the parking speed threshold, and the current spacing does not exceed the second spacing threshold, then the corresponding type of traffic light abnormality is set to no abnormality; if the current road section driving direction is left turn or straight ahead, and the current crossing line state is not crossing the line, and the current vehicle speed has exceeded If the current driving direction of the road section is left turn and the current crossing status is crossed, the corresponding type of signal light abnormality is set to no abnormality; if the current driving direction of the road section is straight ahead and the current crossing status is crossed, the corresponding type of signal light abnormality is set to go straight ahead on red light; if the current driving direction of the road section is right turn and the current vehicle speed does not exceed the stopping speed threshold, the corresponding type of signal light abnormality is set to stop on red light right turn; if the current driving direction of the road section is right turn and the current vehicle speed exceeds the stopping speed threshold, the corresponding type of signal light abnormality is set to no abnormality;

[0144] Here, the second spacing threshold is a preset distance length;

[0145] Step 3-5-6, when the current signal light state is green, if the current vehicle speed has exceeded the parking speed threshold, the corresponding type of signal light abnormality is set to no abnormality; if the current vehicle speed does not exceed the parking speed threshold, the corresponding type of signal light abnormality is set to green light start time too long;

[0146] Step 3-5-7, identifying the obtained type of abnormality of the first type of signal light; if the type of abnormality of the first type of signal light is not no abnormality, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset third time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding fourth time period by the start time of the current time and the end time of the current time, forming a corresponding type of abnormality record of the first type of signal light by the fourth time period obtained this time and the type of abnormality of the first type of signal light, and adding it to the type of abnormality queue of the first type of signal light;

[0147] Here, the third duration is a preset time length;

[0148] Step 3-6, based on the vehicle data queue and the perception data queue, identify whether the vehicle continues to go straight at a known red light, continues to turn left at a known red light, stops at a wrong right turn at a red light, stops too early at a red light resulting in the parking position being too far from the stop line, and takes too long to start the vehicle at a green light when the state of the direction indicator light is known, and update the corresponding second-class signal light abnormality queue based on the identification results;

[0149] Specifically, it includes: step 3-6-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record closest to the current perception record in the vehicle data queue is recorded as the corresponding current vehicle record;

[0150] Step 3-6-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a direction indicator light perception target to obtain a corresponding current confirmation result;

[0151] Among them, the current confirmation results include existence and non-existence;

[0152] Step 3-6-3, identify the current confirmation result; if the current confirmation result is existence, the target road section lane marking in the current perception record matches the road section lane marking recorded by the current vehicle and the perception target of the target type is the direction indicator signal light is recorded as the corresponding current signal light, and the perception target of the target type is the stop line in the current perception record and the closest to the rear axle center coordinates recorded by the current vehicle is recorded as the corresponding current stop line, and the signal light state of the current signal light is recorded as the corresponding current signal light state, and go to step 3-6-4; if the current confirmation result is non-existent, set the corresponding second-category signal light abnormality type to no abnormality and go to step 3-6-7;

[0153] Step 3-6-4, taking the speed recorded by the current vehicle as the corresponding current speed; and confirming whether the road section corresponding to the lane marking of the road section recorded by the current vehicle is a left turn, a right turn or a straight section by querying the high-precision road map to obtain the corresponding current section driving direction; and identifying whether the center coordinates of the vehicle head recorded by the current vehicle have crossed the current stop line, if the center coordinates of the vehicle head have crossed the current stop line, setting the corresponding current crossing state to crossed, if the center coordinates of the vehicle head have not crossed the current stop line, setting the corresponding current crossing state to not crossed, and estimating the shortest distance between the center coordinates of the current vehicle head and the current stop line to obtain the corresponding current distance;

[0154] The current driving direction of the road section includes left turn, right turn and straight ahead; the current crossing status includes crossing the line and not crossing the line;

[0155] Step 3-6-5, when the current traffic light state is red, if the current driving direction of the road section is left turn or straight, and the current crossing state is not crossing the line, and the current vehicle speed does not exceed the parking speed threshold, and the current spacing exceeds the second spacing threshold, then the corresponding second-category traffic light abnormality type is set to red light stop too far from the stop line; if the current driving direction of the road section is left turn or straight, and the current crossing state is not crossing the line, and the current vehicle speed does not exceed the parking speed threshold, and the current spacing does not exceed the second spacing threshold, then the corresponding second-category traffic light abnormality type is set to no abnormality; if the current driving direction of the road section is left turn or straight, and the current crossing state is not crossing the line, and the current vehicle speed has exceeded If the current driving direction is left turn and the current crossing status is crossed, the corresponding second-class signal light abnormality type is set to no abnormality; if the current driving direction is straight ahead and the current crossing status is crossed, the corresponding second-class signal light abnormality type is set to go straight ahead on red light; if the current driving direction is right turn and the current vehicle speed does not exceed the parking speed threshold, the corresponding second-class signal light abnormality type is set to stop on red light right turn; if the current driving direction is right turn and the current vehicle speed exceeds the parking speed threshold, the corresponding second-class signal light abnormality type is set to no abnormality;

[0156] Step 3-6-6, when the current signal light state is green, if the current vehicle speed has exceeded the parking speed threshold, the corresponding second-category signal light abnormality type is set to no abnormality; if the current vehicle speed does not exceed the parking speed threshold, the corresponding second-category signal light abnormality type is set to green light start time too long;

[0157] Step 3-6-7, identify the obtained second-category signal light abnormality type; if the second-category signal light abnormality type is not no abnormality, use the first timestamp of the current vehicle record as the corresponding end time, and subtract a preset fourth time length from the end time as the corresponding start time, and form a corresponding fifth time period by the start time and the end time, and form a corresponding second-category signal light abnormality record by the fifth time period obtained this time and the second-category signal light abnormality type and add it to the second-category signal light abnormality queue;

[0158] Here, the fourth duration is a preset time length;

[0159] Step 3-7, based on the vehicle data queue and the perception data queue, identify whether the vehicle continues to go straight when the lane is known to be prohibited, stops without changing lanes when the lane is prohibited, and stops when the lane is known to be passable, and update the corresponding three types of signal light abnormality queues based on the identification results;

[0160] Specifically, it includes: step 3-7-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record time is recorded as the corresponding current vehicle record;

[0161] Step 3-7-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a lane signal light perception target to obtain the corresponding current confirmation result;

[0162] Among them, the current confirmation results include existence and non-existence;

[0163] Step 3-7-3, identify the current confirmation result; if the current confirmation result is existence, the lane marking of the target road section in the current perception record is matched with the lane marking of the road section recorded by the current vehicle and the perception target of the target type is the lane signal light as the corresponding current signal light, and the signal light state of the current signal light is recorded as the corresponding current signal light state, and the speed and heading angle recorded by the current vehicle are used as the corresponding current speed and current heading angle, and go to step 3-7-4; if the current confirmation result is non-existent, set the corresponding three types of signal light abnormality types to no abnormality and go to step 3-7-6;

[0164] Step 3-7-4, when the current signal light state is no-go, extract the first specified number of vehicle data records before the current vehicle record in the vehicle data queue and form a corresponding current record sequence with the current vehicle record; and based on the current record sequence, confirm whether the vehicle is in the process of changing lanes to obtain the corresponding current lane-changing state; and when the current lane-changing state is changing lanes and the current vehicle speed exceeds the parking speed threshold, set the corresponding three types of signal light abnormality types to no abnormality; and when the current vehicle speed does not exceed the parking speed threshold, set the corresponding three types of signal light abnormality types to no-go and stop without changing lanes; and when the current lane-changing state is not changing lanes and the current vehicle speed exceeds the parking speed threshold, set the corresponding three types of signal light abnormality types to no-go and stop without stopping;

[0165] The first designated number is a preset positive integer; the current lane change state includes lane changing and lane not changing;

[0166] Step 3-7-5, when the current signal light state is pass, identify the current vehicle speed; if the current vehicle speed does not exceed the parking speed threshold, set the corresponding three types of signal light abnormality types to pass parking; if the current vehicle speed exceeds the parking speed threshold, set the corresponding three types of signal light abnormality types to no abnormality;

[0167] Step 3-7-6, identifying the obtained three types of signal light abnormality types; if the three types of signal light abnormality types are not no abnormality, the first timestamp recorded by the current vehicle is used as the corresponding end time of the current time, and the time obtained by subtracting a preset fifth time length from the end time of the current time is used as the corresponding start time of the current time, and the current start time and the current end time form a corresponding sixth time period, and the sixth time period obtained this time and the three types of signal light abnormality types form a corresponding three types of signal light abnormality record and add it to the three types of signal light abnormality queue;

[0168] Here, the fifth duration is a preset time length;

[0169] Step 3-8, identifying whether the vehicle has three abnormal conditions when driving on a curve, namely, driving too fast, driving too slow, and stopping on a curve, according to the vehicle data queue, and updating the corresponding curve abnormality queue based on the identification result;

[0170] Specifically, it includes: step 3-8-1, when a vehicle data record is added to the vehicle data queue, the vehicle data record added at that time is recorded as the corresponding current vehicle record;

[0171] Step 3-8-2, extracting the second specified number of vehicle data records before the current vehicle record in the vehicle data queue to form a corresponding current record sequence with the current vehicle record; and calculating the mean of the road curvature in the current record sequence to obtain the corresponding current average curvature;

[0172] Wherein, the second specified number is a preset positive integer;

[0173] Step 3-8-3, identifying whether the current average curvature exceeds a preset first curvature threshold; if so, setting the corresponding current curve state to curve; if not, setting the corresponding current curve state to non-curve;

[0174] The first curvature threshold is a preset curvature parameter; the current curve state includes curves and non-curves;

[0175] Step 3-8-4, when the current curve state is a curve, the speed and lane speed upper and lower limits recorded by the current vehicle are used as the corresponding current speed and current speed range; and the current speed and current speed range are identified; if the current speed does not exceed the current speed range, the ratio of the current speed to the upper speed threshold of the current speed range is calculated and whether the current ratio exceeds the preset second ratio threshold is identified, if the current ratio exceeds the second ratio threshold, the corresponding curve abnormality type is set to no abnormality, if the current ratio does not exceed the second ratio threshold, the corresponding curve abnormality type is set to curve low speed; if the current speed is lower than the lower speed threshold of the current speed range, the corresponding curve abnormality type is set to curve low speed; if the current speed exceeds the upper speed threshold of the current speed range, the corresponding curve abnormality type is set to curve overspeed; if the current speed does not exceed the parking speed threshold, the corresponding curve abnormality type is set to curve parking;

[0176] Wherein, the second ratio threshold is a preset ratio;

[0177] Step 3-8-5, when the current curve state is non-curve, set the corresponding curve abnormality type to no abnormality;

[0178] Step 3-8-6, identifying the obtained curve abnormality type; if the curve abnormality type is not no abnormality, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset sixth time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding seventh time period by the start time of the current time and the end time of the current time, and forming a corresponding curve abnormality record by the seventh time period obtained this time and the curve abnormality type to add to the curve abnormality queue;

[0179] Here, the sixth duration is a preset time length;

[0180] Step 3-9, identifying whether the vehicle is driving on the ramp at an excessive speed, driving at an excessively low speed, or parking on the ramp according to the vehicle data queue, and updating the corresponding ramp abnormality queue based on the identification result;

[0181] Specifically, it includes: step 3-9-1, when a vehicle data record is added to the vehicle data queue, the vehicle data record added at that time is recorded as the corresponding current vehicle record;

[0182] Step 3-9-2, query the high-precision road map to identify whether the road section corresponding to the road section lane sign recorded by the current vehicle is a ramp section to obtain the corresponding current ramp state;

[0183] Among them, the current ramp status includes ramp and non-ramp;

[0184] Step 3-9-3, when the current ramp state is ramp, the speed and lane speed upper and lower limits recorded by the current vehicle are used as the corresponding current speed and current speed range; and the current speed and current speed range are identified; if the current speed does not exceed the current speed range, the ratio of the current speed to the upper speed threshold of the current speed range is calculated and whether the current ratio exceeds the preset third ratio threshold is identified, if the current ratio exceeds the third ratio threshold, the corresponding ramp abnormality type is set to no abnormality, if the current ratio does not exceed the third ratio threshold, the corresponding ramp abnormality type is set to ramp low speed; if the current speed is lower than the lower speed threshold of the current speed range, the corresponding ramp abnormality type is set to ramp low speed; if the current speed exceeds the upper speed threshold of the current speed range, the corresponding ramp abnormality type is set to ramp speeding; if the current speed does not exceed the parking speed threshold, the corresponding ramp abnormality type is set to ramp parking;

[0185] Wherein, the third ratio threshold is a preset ratio;

[0186] Step 3-9-4, when the current ramp state is non-ramp, set the corresponding ramp abnormality type to no abnormality;

[0187] Step 3-9-5, identifying the obtained ramp anomaly type; if the ramp anomaly type is not no anomaly, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset seventh time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding eighth time period by the start time of the current time and the end time of the current time, and forming a corresponding ramp anomaly record by the eighth time period obtained this time and the ramp anomaly type to add to the ramp anomaly queue;

[0188] Here, the seventh duration is a preset time length;

[0189] Step 3-10, identifying whether the vehicle has two abnormal conditions, i.e., lane changing and parking in the tunnel, when driving in the tunnel according to the vehicle data queue, and updating the corresponding tunnel abnormality queue based on the identification result;

[0190] Specifically, it includes: step 3-10-1, when a vehicle data record is added to the vehicle data queue, the vehicle data record added at that time is recorded as the corresponding current vehicle record;

[0191] Step 3-10-2, query the high-precision road map to identify whether the road section corresponding to the lane mark of the road section recorded by the current vehicle is a tunnel section to obtain the corresponding current tunnel state; and use the speed recorded by the current vehicle as the corresponding current speed;

[0192] Among them, the current tunnel status includes tunnel and non-tunnel;

[0193] Step 3-10-3, when the current tunnel state is tunnel, extract the third specified number of vehicle data records before the current vehicle record in the vehicle data queue to form a corresponding current record sequence with the current vehicle record; and based on the current record sequence, confirm whether the vehicle is in the process of changing lanes to obtain the corresponding current lane changing state; and identify the current lane changing state and the current vehicle speed; if the current lane changing state is changing lanes, set the corresponding tunnel abnormality type to tunnel lane changing; if the current lane changing state is not changing lanes, identify whether the current vehicle speed exceeds the parking speed threshold, if it exceeds, set the corresponding tunnel abnormality type to tunnel parking, if it has exceeded, set the corresponding tunnel abnormality type to no abnormality;

[0194] Wherein, the third specified number is a preset positive integer;

[0195] Step 3-10-4, when the current tunnel state is non-tunnel, set the corresponding tunnel exception type to no exception;

[0196] Step 3-10-5, identifying the obtained tunnel anomaly type; if the tunnel anomaly type is not no anomaly, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset eighth time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding ninth time period by the start time and the end time of the current time, and forming a corresponding tunnel anomaly record by the ninth time period obtained this time and the tunnel anomaly type and adding it to the tunnel anomaly queue;

[0197] Here, the eighth duration is a preset time length;

[0198] Step 3-11, identifying whether the vehicle has two abnormal conditions, namely, lowering the bar without deceleration and raising the bar without acceleration, at the entrance or exit of a toll station or parking lot with a lifting bar, based on the vehicle data queue and the perception data queue, and updating the corresponding lifting bar abnormality queue based on the identification result;

[0199] Specifically, it includes: step 3-11-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record time is recorded as the corresponding current vehicle record;

[0200] Step 3-11-2, confirming whether there is a target road section lane sign in the current perception record that matches the road section lane sign recorded by the current vehicle and the target type is a lift pole to obtain a corresponding current confirmation result;

[0201] Among them, the current confirmation results include existence and non-existence;

[0202] Step 3-11-3, identify the current confirmation result; if the current confirmation result is existence, match the target road section lane sign in the current perception record with the road section lane sign recorded by the current vehicle and record the perception target of the target type as a lift pole as the corresponding current lift pole, and record the lifting state of the current lift pole as the corresponding current lifting state, and go to step 3-11-4; if the current confirmation result is non-existent, set the corresponding lift pole abnormality type to no abnormality and go to step 3-11-5;

[0203] Step 3-11-4, identify the current lifting state; if the current lifting state is raising the bar, identify the speed recorded by the current vehicle, if the current speed exceeds the parking speed threshold, set the corresponding lifting bar abnormality type to no abnormality, if the current speed does not exceed the parking speed threshold, set the corresponding lifting bar abnormality type to raising the bar and parking; if the current lifting state is lowering the bar, identify the speed recorded by the current vehicle, if the current speed exceeds the parking speed threshold, set the corresponding lifting bar abnormality type to lowering the bar without parking, if the current speed does not exceed the parking speed threshold, set the corresponding lifting bar abnormality type to no abnormality;

[0204] Step 3-11-5, identifying the obtained lift lever abnormality type; if the lift lever abnormality type is not no abnormality, taking the first timestamp of the current vehicle record as the corresponding end time of the current time, and subtracting a preset ninth time length from the end time of the current time as the corresponding start time of the current time, and forming a corresponding tenth time period by the start time of the current time and the end time of the current time, and forming a corresponding lift lever abnormality record by the tenth time period obtained this time and the lift lever abnormality type to add to the lift lever abnormality queue;

[0205] Here, the ninth duration is a preset time length;

[0206] Step 3-12, based on the vehicle data queue and the perception data queue, identify five abnormal conditions when the vehicle is driving on the lane, including: occupying the lane during non-traffic hours, not detouring around stationary vehicles, not detouring around obstacles, colliding with other moving vehicles around the vehicle, and parking on the lane, and update the corresponding lane abnormality queue based on the identification results;

[0207] Specifically, it includes: step 3-12-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record time is recorded as the corresponding current vehicle record;

[0208] Step 3-12-2, identifying whether the speed recorded by the current vehicle is lower than the parking speed threshold; if so, setting the corresponding lane abnormality type to abnormal parking; if not, setting the corresponding lane abnormality type to no abnormality;

[0209] Step 3-12-3, by querying the high-precision road map, the passable time period of the road section corresponding to the lane mark of the road section recorded by the current vehicle is used as the corresponding current passable time period; and whether the first timestamp recorded by the current vehicle meets the current passable time period is identified; if not, the corresponding lane abnormality type is set to passable during non-passable time periods; if satisfied, the corresponding lane abnormality type is set to no abnormality;

[0210] Step 3-12-4, by querying the high-precision road map, identify the map coverage area of ​​each perception target whose target type is a vehicle and whose speed is lower than the parking speed threshold in the current perception record to obtain the corresponding stationary vehicle range; and identify the map coverage area of ​​each perception target whose target type is an obstacle and whose speed is lower than the parking speed threshold in the current perception record to obtain the corresponding obstacle range; and identify the map coverage area of ​​each perception target whose target type is a vehicle and whose speed is not lower than the parking speed threshold in the current perception record to obtain the corresponding moving vehicle range; and identify the map coverage area of ​​the own vehicle based on the current vehicle record to obtain the corresponding current own vehicle range; and identify whether each stationary vehicle range has an intersection with the current own vehicle range, if at least one stationary vehicle range has an intersection with the current own vehicle range If there is no intersection between the ranges of all stationary vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "no detour obstacle". If there is no intersection between the ranges of all stationary vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "no detour obstacle". If there is no intersection between the ranges of all obstacle vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "no detour obstacle". If there is no intersection between the ranges of all obstacle vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "no intersection". If there is no intersection between the ranges of all moving vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "vehicle collision". If there is no intersection between the ranges of all moving vehicles and the current self-vehicle range, the corresponding lane anomaly type is set to "no exception".

[0211] Step 3-12-5, identifying the lane anomaly type obtained; if the lane anomaly type is not no anomaly, taking the first timestamp of the current vehicle record as the corresponding end time, and subtracting a preset tenth time period from the end time as the corresponding start time, and forming a corresponding eleventh time period by the start time and the end time, and forming a corresponding lane anomaly record by the eleventh time period obtained this time and the lane anomaly type and adding it to the lane anomaly queue;

[0212] Here, the tenth duration is a preset time length;

[0213] Step 3-13, based on the vehicle data queue and the perception data queue, identify whether the vehicle fails to yield to the left straight vehicle during the process of turning right, fails to yield to the right straight vehicle during the process of turning left, and fails to yield to the right left-turning or straight-moving vehicle during the process of going straight, and update the corresponding no-light intersection abnormal queue based on the identification results;

[0214] Specifically, it includes: step 3-13-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record time is recorded as the corresponding current vehicle record;

[0215] Step 3-13-2, if the traffic intersection type recorded by the current vehicle is a meeting intersection, then by querying the high-precision road map, whether the intersection where the rear axle center coordinates of the current vehicle are located is a no-light intersection to obtain the corresponding current intersection recognition state;

[0216] Among them, the current intersection recognition status includes unlit intersection and non-unlit intersection;

[0217] Step 3-13-3, identify the current intersection recognition state; if the current intersection recognition state is an unlit intersection, go to step 3-13-4; if the current intersection recognition state is a non-unlit intersection, set the corresponding unlit intersection abnormality type to no abnormality and go to step 3-13-8;

[0218] Step 3-13-4, confirming the driving direction of the vehicle at the current intersection based on the vehicle data queue to obtain the corresponding current vehicle driving direction;

[0219] Among them, the current driving direction includes left turn, straight ahead and right turn;

[0220] Step 3-13-5, when the current direction of the vehicle is to turn left, identify whether there is a straight-moving vehicle on the right side of the current intersection according to the perception data queue; if there is a straight-moving vehicle on the right side, identify whether the vehicle slows down and gives way to the straight-moving vehicle on the right side; if the vehicle does not slow down and give way, set the corresponding no-light intersection exception type to left turn without giving way to the right straight-moving vehicle; if the vehicle slows down and gives way, set the corresponding no-light intersection exception type to no exception;

[0221] Step 3-13-6, when the current vehicle direction is straight ahead, identify whether there is a left-turning or straight-going vehicle on the right side of the current intersection according to the perception data queue; if there is a left-turning or straight-going vehicle on the right side of the vehicle, identify whether the vehicle slows down to give way to the vehicle on the right; if the vehicle does not slow down to give way, set the corresponding no-light intersection exception type to straight ahead without giving way to the right left turn or right straight ahead; if the vehicle slows down to give way, set the corresponding no-light intersection exception type to no exception;

[0222] Step 3-13-7, when the current direction of the vehicle is a right turn, identify whether there is a straight-moving vehicle on the left side of the current intersection according to the perception data queue; if there is a straight-moving vehicle on the left side of the vehicle, identify whether the vehicle slows down to give way to the vehicle on the left; if the vehicle does not slow down to give way, set the corresponding no-light intersection exception type to right turn without giving way to the left straight-moving vehicle; if the vehicle slows down to give way, set the corresponding no-light intersection exception type to no exception;

[0223] Step 3-13-8, identifying the obtained abnormal type of the unlighted intersection; if the abnormal type of the unlighted intersection is not no abnormality, the first timestamp of the current vehicle record is used as the corresponding end time of the current time, and the time obtained by subtracting a preset eleventh time period from the end time of the current time is used as the corresponding start time of the current time, and the start time of the current time and the end time of the current time form a corresponding twelfth time period, and the twelfth time period obtained this time and the abnormal type of the unlighted intersection form a corresponding unlighted intersection abnormal record and add it to the unlighted intersection abnormal queue;

[0224] Here, the eleventh duration is a preset time length;

[0225] Step 3-14, identifying three abnormal conditions when the vehicle is at a roundabout, including failure to exit according to a planned exit, emergency braking / stopping when entering the intersection, and emergency braking / stopping when exiting the intersection, based on the vehicle data queue and the perception data queue, and updating the corresponding roundabout abnormality queue based on the identification results;

[0226] Specifically, it includes: step 3-14-1, when a perception data record is added to the perception data queue, the perception data record added at that time is recorded as the corresponding current perception record; and the vehicle data record in the vehicle data queue that is closest to the current perception record time is recorded as the corresponding current vehicle record;

[0227] Step 3-14-2, if the traffic intersection type recorded by the current vehicle is a meeting intersection, then by querying the high-precision road map, whether the intersection where the rear axle center coordinates recorded by the current vehicle are located is identified as a roundabout to obtain the corresponding current intersection identification state;

[0228] Among them, the current intersection recognition status includes roundabouts and non-roundabouts;

[0229] Step 3-14-3, identify the current intersection recognition state; if the current intersection recognition state is a roundabout, go to step 3-14-4; if the current intersection recognition state is a non-roundabout, set the corresponding roundabout exception type to no exception and go to step 3-14-8;

[0230] Step 3-14-4, the speed and rear axle center coordinates recorded by the current vehicle are marked as the corresponding current speed and current position of the vehicle; and the planned trajectory output by the trajectory planning module of the vehicle system confirms the entrance and exit positions of the vehicle at the current intersection to obtain the corresponding planned entrance position and planned exit position; and when the current position of the vehicle matches the planned entrance position, the corresponding current position recognition state is set to the entrance, and when the current position of the vehicle matches the planned exit position, the corresponding current position recognition state is set to the exit, and when the current position of the vehicle does not match the planned entrance position and the planned exit position, the corresponding current position recognition state is set to the inside of the intersection;

[0231] The current location recognition status includes entrance, exit, and intersection;

[0232] Step 3-14-5, when the current position recognition state is the entrance, identify whether the current vehicle speed is lower than the parking speed threshold; if so, set the corresponding roundabout abnormality type to emergency braking or parking at the intersection; if not, set the corresponding roundabout abnormality type to no abnormality;

[0233] Step 3-14-6, when the current position recognition state is an exit, identify whether the current vehicle speed is lower than the parking speed threshold; if so, set the corresponding roundabout abnormality type to emergency braking or parking at the exit; if not, set the corresponding roundabout abnormality type to no abnormality;

[0234] Step 3-14-7, when the current position recognition state is within the intersection, identify whether the current vehicle position has missed the planned exit position; if so, set the corresponding roundabout exception type to not turning out according to the planned exit; if not, set the corresponding roundabout exception type to no exception;

[0235] Step 3-14-8, identifying the obtained roundabout abnormality type; if the roundabout abnormality type is not no abnormality, taking the first timestamp of the current vehicle record as the corresponding end time, and subtracting a preset twelfth time from the end time as the corresponding start time, and forming a corresponding thirteenth time period by the start time and the end time, and forming a corresponding roundabout abnormality record by the thirteenth time period obtained this time and the roundabout abnormality type and adding it to the roundabout abnormality queue;

[0236] Here, the twelfth duration is a preset time length.

[0237] Step 4: After the first vehicle is finished evaluating, a comprehensive score is calculated based on the fourteen types of abnormal queues to obtain a corresponding first evaluation report;

[0238] Specifically, step 41 is to count the number of each type of abnormality in the fourteen types of abnormality queues to obtain the corresponding number of fourteen types of abnormalities;

[0239] Among them, the 14 types of abnormal numbers include the number of speed limit abnormalities, the number of U-turn abnormalities, the number of main and auxiliary road switching abnormalities, the number of stop and yield abnormalities, the number of type I traffic lights abnormalities, the number of type II traffic lights abnormalities, the number of type III traffic lights abnormalities, the number of curve abnormalities, the number of ramp abnormalities, the number of tunnel abnormalities, the number of lifting pole abnormalities, the number of lane abnormalities, the number of unlit intersection abnormalities and the number of roundabout abnormalities;

[0240] Step 42, and record the abnormal type corresponding to each abnormal number that is not 0 in the fourteen types of abnormal numbers as a corresponding abnormal evaluation type;

[0241] Step 43, based on the preset first correspondence table, the abnormal deduction score corresponding to each abnormal evaluation type is identified to obtain the corresponding first deduction score;

[0242] Here, the first correspondence table is a correspondence table reflecting the correspondence between the abnormality type and the abnormality deduction value; the first correspondence table includes fourteen first correspondence records; the first correspondence record includes a first abnormality type field and a first deduction value field; the first abnormality type field includes a speed limit abnormality type, a U-turn abnormality type, a main and auxiliary road switching abnormality type, a stop and yield abnormality type, a first type of signal light abnormality type, a second type of signal light abnormality type, a third type of signal light abnormality type, a curve abnormality type, a ramp abnormality type, a tunnel abnormality type, a lift pole abnormality type, a lane abnormality type, an unlit intersection abnormality type, and a roundabout abnormality type; the first deduction value field is an integer score;

[0243] Step 44, sum up all the first deduction values ​​obtained to obtain the corresponding total deduction value; and calculate the corresponding evaluation score for the time = full score - total deduction value based on the preset full score and total deduction value; and form a corresponding first evaluation record by obtaining each abnormal evaluation type and the corresponding first deduction value; and form a corresponding first evaluation report by all the first evaluation records obtained and the evaluation score for the time.

[0244] Figure 4 The module structure diagram of a processing device for evaluating the perception response performance of a vehicle provided in the second embodiment of the present invention, the device is a terminal device or a server that implements the aforementioned method embodiment, and can also be a device that enables the aforementioned terminal device or server to implement the aforementioned method embodiment, for example, the device can be a device or chip system of the aforementioned terminal device or server. Figure 4 As shown, the processing device 200 for evaluating the perception response performance of a vehicle provided in the second embodiment of the present invention is connected to the on-board system of the first vehicle 100; the processing device 200 includes: a data acquisition module 2001, a data storage module 2002, an abnormality identification module 2003 and an abnormality evaluation module 2004.

[0245] The data collection module 2001 is used to initialize two empty data collection queues as corresponding vehicle data queues and perception data queues and store them in the data storage module 2002 when the first vehicle starts the evaluation; and initialize fourteen empty exception queues to form corresponding fourteen types of exception queues and store them in the data storage module 2002; the fourteen types of exception queues include speed limit exception queue, U-turn exception queue, main and auxiliary road switching exception queue, stop and give way exception queue, type I traffic light exception queue, type II traffic light exception queue, type III traffic light exception queue, curve exception queue, ramp exception queue, tunnel exception queue, lift bar exception queue, lane exception queue, unlit intersection exception queue and roundabout exception queue.

[0246] The data acquisition module 2001 is also used to collect data on the real-time motion, positioning and road information output by the vehicle-mounted system during the evaluation of the first vehicle, generate corresponding vehicle data records and add them to the vehicle data queue; and collect data on the target perception data output by the vehicle-mounted system, generate corresponding perception data records and add them to the perception data queue;

[0247] The anomaly recognition module 2003 is used to perform fourteen types of perception response anomaly recognition and processing according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues during the evaluation process of the first vehicle; the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and give way anomalies, Class I traffic light anomalies, Class II traffic light anomalies, Class III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies and roundabout anomalies.

[0248] The abnormal evaluation module 2004 is used to obtain a corresponding first evaluation report by comprehensively scoring the fourteen types of abnormal queues after the first vehicle completes the evaluation.

[0249] An embodiment of the present invention provides a processing device for evaluating the perception response performance of a vehicle, which can execute the method steps in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here.

[0250] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the data acquisition module can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The function of the above-mentioned module is determined. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element described here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0251] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital signal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a module above is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call program code. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0252] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the above method embodiments are generated. The above computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The above-mentioned computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the above-mentioned computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, Bluetooth, microwave, etc.) methods. The above-mentioned computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The above-mentioned available medium can be a magnetic medium (such as a floppy disk, a hard disk, a tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0253] Figure 5 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention. The electronic device may be a terminal device or a server for implementing the method of the aforementioned embodiment, or may be a terminal device or a server for implementing the method of the aforementioned embodiment connected to the aforementioned terminal device or server. Figure 5As shown, the electronic device may include: a processor 301 (such as a CPU), a memory 302, and a transceiver 303; the transceiver 303 is coupled to the processor 301, and the processor 301 controls the transceiver 303. Various instructions may be stored in the memory 302 to complete various processing functions and implement the processing steps described in the aforementioned embodiment method. Preferably, the electronic device involved in the embodiment of the present invention also includes: a power supply 304, a system bus 305 and a communication port 306. The system bus 305 is used to realize the communication connection between components. The above-mentioned communication port 306 is used for connecting and communicating between the electronic device and other peripherals.

[0254] exist Figure 5 The system bus 305 mentioned in the figure can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used to realize the communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM), and may also include non-volatile memory (Non-Volatile Memory), such as at least one disk storage.

[0255] The above-mentioned processor can be a general-purpose processor, including a central processing unit CPU, a network processor (Network Processor, NP), a graphics processing unit (Graphics Processing Unit, GPU), etc.; it can also be a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0256] It should be noted that an embodiment of the present invention further provides a computer-readable storage medium, in which instructions are stored. When the computer-readable storage medium is run on a computer, the computer executes the method and processing process provided in the above embodiments.

[0257] The embodiment of the present invention provides a processing method, device, electronic device and computer-readable storage medium for evaluating the perception response performance of a vehicle. It can be seen from the above invention content that when the first vehicle starts the evaluation, the embodiment of the present invention initializes two empty data collection queues as corresponding vehicle data queues and perception data queues, and initializes fourteen empty exception queues to form corresponding fourteen types of exception queues; and during the evaluation of the first vehicle, the real-time motion, positioning and road information output by the vehicle system is collected to generate corresponding vehicle data records and added to the vehicle data queue, and the target perception data output by the vehicle system is collected to generate corresponding perception data records and added to the perception data queue; and during the evaluation of the first vehicle, fourteen types of perception response anomaly recognition processing are performed according to the vehicle data queue, perception data queue and fourteen types of exception queues; and after the first vehicle ends the evaluation, the fourteen types of exception queues are comprehensively scored to obtain the corresponding first evaluation report. The embodiment of the present invention not only gets rid of the manual operation process, but also realizes real-time evaluation during vehicle driving, thereby improving the evaluation efficiency and the real-time performance of the evaluation.

[0258] The professionals should further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0259] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0260] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for evaluating the perception response performance of a vehicle, characterized in that: The method comprises: When the first vehicle starts the evaluation, two empty data collection queues are initialized and recorded as the corresponding vehicle data queue and perception data queue; and fourteen empty exception queues are initialized to form corresponding fourteen types of exception queues; the fourteen types of exception queues include speed limit exception queue, U-turn exception queue, main and auxiliary road switching exception queue, stop and yield exception queue, first type signal light exception queue, second type signal light exception queue, third type signal light exception queue, curve exception queue, ramp exception queue, tunnel exception queue, lift pole exception queue, lane exception queue, no light intersection exception queue and roundabout exception queue; During the evaluation of the first vehicle, data collection is performed on the real-time motion, positioning and road information output by the vehicle system to generate corresponding vehicle data records and add them to the vehicle data queue; and data collection is performed on the target perception data output by the vehicle system to generate corresponding perception data records and add them to the perception data queue; During the evaluation of the first vehicle, fourteen types of perception response anomaly identification processing are performed according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues; the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and yield anomalies, type I traffic light anomalies, type II traffic light anomalies, type III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies and roundabout anomalies; After the first vehicle is evaluated, a comprehensive score is calculated based on the fourteen types of abnormal queues to obtain a corresponding first evaluation report.

2. The method for evaluating the perception response performance of a vehicle according to claim 1, characterized in that: The vehicle data queue is composed of a plurality of vehicle data records; the vehicle data record is composed of a corresponding first timestamp, vehicle speed, acceleration, heading angle, vehicle head center coordinates, rear axle center coordinates, road curvature, road section lane markings, traffic intersection type, and lane speed upper and lower limits; the traffic intersection type includes straight road / meeting intersection; The vehicle speed, the acceleration, the heading angle, the center coordinates of the front of the vehicle, the center coordinates of the rear axle, the road curvature, the lane sign, the traffic intersection type and the upper and lower limits of lane speed are the real-time driving speed of the vehicle, the real-time acceleration of the vehicle, the real-time heading angle of the vehicle, the real-time coordinates of the center point of the top of the front of the vehicle, the real-time coordinates of the center point of the rear axle of the vehicle, the curvature information of the corresponding point on the high-precision map of the intersection of the center point of the rear axle of the vehicle and the vertical line of the road surface, the road sign + section sign + lane sign corresponding to the lane where the vehicle is located, the classification information of the traffic intersection where the vehicle is located, and the upper and lower speed limit thresholds of the lane where the vehicle is located.

3. The method for evaluating the perception response performance of a vehicle according to claim 2, characterized in that: The perception data queue is composed of a plurality of the perception data records; the perception data record is composed of a corresponding second timestamp and a plurality of perception targets; the perception target is composed of a corresponding target identifier, target type and target attribute; The target identifier is a unique identifier assigned to each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module; The target type is the identification object type given for each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module, including at least vehicles, obstacles, lifting poles, U-turn signs, road entrance and exit signs, speed limit signs, stop and yield signs, motor vehicle signal lights, direction indicator lights, lane signal lights, stop and yield lines, and stop lines; The target attribute is a set of object attributes given to each target when the vehicle-mounted system identifies the targets around the vehicle body through the built-in perception module, and is composed of two categories: general attributes and individual attributes; the general attributes are common attributes possessed by each target type; the individual attributes are personalized attributes unique to some target types; The general attributes include coordinates, speed, direction, target frame vertex set, target frame size and target road section lane identification; the coordinates, speed, direction, target frame vertex set and target frame size are the position coordinates of the target ground center point, the relative motion speed between the target and the vehicle, the relative motion direction between the target and the vehicle, the vertex coordinate set of the target corresponding detection frame and the geometric size of the target corresponding detection frame given for each target when the vehicle-mounted system detects and identifies the targets around the vehicle body through the built-in perception module; the target road section lane identification is the road identification + section identification + lane identification of the target corresponding lane given when the vehicle-mounted system refreshes the corresponding lane information of each target around the vehicle body through the built-in perception module and the high-precision map module; When the target type is a vehicle, an obstacle, a U-turn sign, a stop sign, a stop line or a stop line, the corresponding individual attribute is empty; When the target type is a lifting rod, the corresponding individual attributes include lifting status; The lifting state includes raising the rod and lowering the rod; When the target type is a road entrance and exit sign, the corresponding individual attribute includes the entrance and exit type; the entrance and exit type includes exit and entrance; When the target type is a speed limit sign, the corresponding individual attributes include the start and end speed limit type and the start and end speed limit value; the start and end speed limit type includes start and end; the start and end speed limit value corresponds to the upper speed limit threshold; When the target type is a motor vehicle signal light, a direction indicator light or a lane signal light, the corresponding individual attributes include the signal light status; the signal light status corresponding to the motor vehicle signal light and the direction indicator light includes red light and green light; the signal light status corresponding to the lane signal light includes pass and no pass.

4. The method for evaluating the perception response performance of a vehicle according to claim 3, characterized in that: When the speed limit exception queue is not empty, it is composed of multiple speed limit exception records; each speed limit exception record corresponds to a speed limit exception event; the speed limit exception record is composed of a corresponding first moment and a speed limit exception type; the first moment is the time information corresponding to the current speed limit exception event; The speed limit abnormality type is the type information corresponding to the speed limit abnormality event, including exceeding the speed limit and speed being too low; When the U-turn abnormality queue is not empty, it is composed of a plurality of U-turn abnormality records; each of the U-turn abnormality records corresponds to a U-turn abnormality event; the U-turn abnormality record is composed of a corresponding first time period and a U-turn abnormality type; the first time period is the time period information corresponding to the current U-turn abnormality event; the U-turn abnormality type is the type information corresponding to the current U-turn abnormality event, including excessive speed and excessive time; When the main-auxiliary road switching abnormality queue is not empty, it is composed of a plurality of main-auxiliary road switching abnormality records; each of the main-auxiliary road switching abnormality records corresponds to a main-auxiliary road switching abnormality event; The main-auxiliary road switching abnormal record is composed of the corresponding second time period and the main-auxiliary road switching abnormal type; the second time period is the time period information corresponding to the main-auxiliary road switching abnormal event; The main-auxiliary road switching abnormal type is the type information corresponding to the main-auxiliary road switching abnormal event, including main-auxiliary road output without deceleration and auxiliary-main road input without acceleration; When the stop and yield abnormality queue is not empty, it is composed of a plurality of stop and yield abnormality records; each of the stop and yield abnormality records corresponds to a stop and yield abnormality event; the stop and yield abnormality record is composed of a corresponding third time period and a stop and yield abnormality type; the third time period is the time period information corresponding to the stop and yield abnormality event; the stop and yield abnormality type is the type information corresponding to the stop and yield abnormality event, including stop and yield crossing the line, being too far from the stop and yield line, and parking for too long; When the first-class signal light abnormality queue is not empty, it is composed of a plurality of first-class signal light abnormality records; each of the first-class signal light abnormality records corresponds to an abnormal event of a motor vehicle signal light; The first type of signal light abnormality record is composed of a corresponding fourth time period and a first type of signal light abnormality type; the fourth time period is the time period information corresponding to the current motor vehicle signal light abnormality event; the first type of signal light abnormality type is the type information corresponding to the current motor vehicle signal light abnormality event, including going straight at a red light, turning left at a red light without stopping, turning right at a red light and stopping, stopping at a red light too far from the stop line, and starting time of a green light too long; When the second-category signal light abnormality queue is not empty, it is composed of a plurality of second-category signal light abnormality records; each of the second-category signal light abnormality records corresponds to a direction indicator signal light abnormality event; the second-category signal light abnormality record is composed of the corresponding fifth time period and the second-category signal light abnormality type; the fifth time period is the time period information corresponding to the current direction indicator signal light abnormality event; the second-category signal light abnormality type is the type information corresponding to the current direction indicator signal light abnormality event, including going straight at a red light, turning left at a red light without stopping, turning right at a red light and stopping, stopping at a red light too far from the stop line, and starting time of a green light too long; When the three types of signal light abnormality queues are not empty, they are composed of a plurality of three types of signal light abnormality records; each of the three types of signal light abnormality records corresponds to a lane signal light abnormality event; The three types of signal light abnormality records are composed of the corresponding sixth time period and the three types of signal light abnormality types; the sixth time period is the time period information corresponding to the lane signal light abnormality event; The three types of signal light abnormality are type information corresponding to the abnormal event of the lane signal light, including no-go without stopping, no-go without changing lanes and stop-and-go; When the curve abnormality queue is not empty, it is composed of a plurality of curve abnormality records; each of the curve abnormality records corresponds to a curve abnormality event; the curve abnormality record is composed of the corresponding seventh time period and the curve abnormality type; the seventh time period is the time period information corresponding to the current curve abnormality event; The curve abnormality type is the type information corresponding to the curve abnormality event, including overspeeding on a curve, underspeed on a curve, and parking on a curve; When the ramp abnormality queue is not empty, it is composed of a plurality of ramp abnormality records; each ramp abnormality record corresponds to a ramp abnormality event; the ramp abnormality record is composed of a corresponding eighth time period and a ramp abnormality type; the eighth time period is the time period information corresponding to the current ramp abnormality event; The ramp abnormality type is the type information corresponding to the ramp abnormality event, including ramp overspeeding, ramp underspeed and ramp parking; When the tunnel exception queue is not empty, it is composed of multiple tunnel exception records; each of the tunnel exception records corresponds to a tunnel exception event; The tunnel anomaly record consists of the corresponding ninth time period and the tunnel anomaly type; The ninth time period is the time period information corresponding to the abnormal tunnel event; The tunnel abnormality type is the type information corresponding to the tunnel abnormality event, including tunnel lane change and tunnel parking; When the lift rod abnormality queue is not empty, it is composed of a plurality of lift rod abnormality records; each lift rod abnormality record corresponds to a lift rod abnormality event; The lifting rod abnormality record is composed of the corresponding tenth time period and the lifting rod abnormality type; the tenth time period is the time period information corresponding to the lifting rod abnormal event; The lift rod abnormality type is the type information corresponding to the lift rod abnormality event, including lowering the rod without stopping and raising the rod stopping; When the lane abnormality queue is not empty, it is composed of a plurality of lane abnormality records; each lane abnormality record corresponds to a lane abnormality event; The lane abnormality record consists of a corresponding eleventh time period and a lane abnormality type; the eleventh time period is the time period information corresponding to the lane abnormality event; The lane abnormality type is the type information corresponding to the lane abnormality event, including passing during non-passing hours, not bypassing stationary vehicles, not bypassing obstacles, vehicle collision and abnormal parking; When the no-light intersection abnormality queue is not empty, it is composed of a plurality of no-light intersection abnormality records; each of the no-light intersection abnormality records corresponds to a no-light intersection abnormality event; The unlighted intersection abnormal record consists of the corresponding twelfth time period and the unlighted intersection abnormal type; the twelfth time period is the time period information corresponding to the unlighted intersection event; The abnormal type of the unlit intersection is the type information corresponding to the abnormal event of the unlit intersection, including right turn without giving way to the left straight, left turn without giving way to the right straight, and straight driving without giving way to the right left turn or right straight; When the roundabout exception queue is not empty, it is composed of a plurality of roundabout exception records; each of the roundabout exception records corresponds to a roundabout exception event; the roundabout exception record is composed of a corresponding thirteenth time period and a roundabout exception type; the thirteenth time period is the time period information corresponding to the current roundabout event; The roundabout abnormality type is type information corresponding to the roundabout abnormal event at the time, including failure to exit according to the planned exit, emergency braking or stopping when entering the intersection, and emergency braking or stopping when exiting the intersection.

5. The method for evaluating the perception response performance of a vehicle according to claim 4, characterized in that: The performing fourteen types of perception response anomaly identification processing according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues specifically includes: Step 501, identifying whether the vehicle has two abnormal conditions, namely, driving too fast or driving too slow, in a known lane speed limit range according to the vehicle data queue and the perception data queue, and updating the corresponding speed limit abnormality queue based on the identification result; Step 502, identifying whether the vehicle has two abnormal conditions, namely, a too fast turning speed and a too long turning time, during a U-turn process according to the vehicle data queue and the perception data queue, and updating the corresponding U-turn abnormality queue based on the identification result; Step 503, identifying whether the vehicle has two abnormal conditions, namely, not decelerating when entering the auxiliary road and not accelerating when entering the main road, during the main-auxiliary road switching process according to the vehicle data queue and the perception data queue, and updating the corresponding main-auxiliary road switching abnormality queue based on the identification result; Step 504, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle has any of the following three abnormal conditions during the stop and yield process: failure to stop in time resulting in crossing the line, premature stopping resulting in the parking position being too far from the stop and yield line, and parking time being too long, and updating the corresponding stop and yield abnormality queue based on the identification results; Step 505, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight at a known red light, continues to turn left at a known red light, stops at a wrong right turn at a red light, stops too early at a red light resulting in the parking position being too far from the stop line, and takes too long to start the vehicle at a green light when the vehicle signal light status is known, and updating the corresponding signal light abnormality queue according to the identification result; Step 506, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight when the direction indicator light is known, continues to turn left when the direction indicator light is known, stops when turning right at a red light, stops too early at a red light so that the parking position is too far from the stop line, and takes too long to start the vehicle when the direction indicator light is known, and updating the corresponding two types of signal light abnormality queues based on the identification results; Step 507, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle continues to go straight when the lane is known to be prohibited, stops without changing lanes when the lane is prohibited, and stops when the lane is known to be passable when the vehicle is stopped when the lane is known to be passable under three abnormal conditions, and updating the corresponding three types of abnormal signal light queues based on the identification results; Step 508, identifying whether the vehicle has three abnormal conditions, namely, excessive speed, low speed, and parking on a curve, when driving on a curve according to the vehicle data queue, and updating the corresponding curve abnormality queue based on the identification result; Step 509, identifying whether the vehicle is driving on the ramp at an excessively high speed, driving at an excessively low speed, or parking on the ramp according to the vehicle data queue, and updating the corresponding ramp abnormality queue based on the identification result; Step 510, identifying whether two abnormal conditions, namely, lane changing and parking in the tunnel, occur when the vehicle is traveling in the tunnel according to the vehicle data queue, and updating the corresponding tunnel abnormality queue based on the identification result; Step 511, identifying whether the vehicle has two abnormal conditions, namely, lowering the bar without deceleration and raising the bar without acceleration, when passing through a toll station entrance or a parking lot entrance with a lifting bar, based on the vehicle data queue and the perception data queue, and updating the corresponding lifting bar abnormality queue based on the identification result; Step 512, identifying five abnormal conditions when the vehicle is traveling on the lane, including occupying the lane during non-traffic hours, not detouring around stationary vehicles, not detouring around obstacles, colliding with other moving vehicles around the vehicle, and parking on the lane, according to the vehicle data queue and the perception data queue, and updating the corresponding lane abnormality queue based on the identification results; Step 513, identifying, based on the vehicle data queue and the perception data queue, whether the vehicle fails to yield to a left straight vehicle during a right turn, fails to yield to a right straight vehicle during a left turn, or fails to yield to a right left-turning or straight vehicle during a straight drive when the vehicle is at a no-light intersection, and updating the corresponding no-light intersection abnormality queue based on the identification results; Step 514, based on the vehicle data queue and the perception data queue, identify three types of abnormal conditions when the vehicle is at a roundabout: the vehicle fails to turn out according to the planned exit, the vehicle emergency brakes / stops when entering the intersection, and the vehicle emergency brakes / stops when turning out of the intersection, and updates the corresponding roundabout abnormal queue based on the identification results.

6. The method for evaluating the perception response performance of a vehicle according to claim 5, characterized in that: The first evaluation report corresponding to the fourteen types of abnormal queues is obtained by comprehensive scoring, specifically including: The number of each type of abnormality of the fourteen types of abnormal queues is counted to obtain the corresponding number of fourteen types of abnormalities; the fourteen types of abnormalities include the number of speed limit abnormalities, the number of U-turn abnormalities, the number of main and auxiliary road switching abnormalities, the number of stop and yield abnormalities, the number of type I traffic lights abnormalities, the number of type II traffic lights abnormalities, the number of type III traffic lights abnormalities, the number of curve abnormalities, the number of ramp abnormalities, the number of tunnel abnormalities, the number of lifting pole abnormalities, the number of lane abnormalities, the number of unlit intersection abnormalities and the number of roundabout abnormalities; And the abnormal type corresponding to each abnormal number that is not 0 in the fourteen types of abnormal numbers is recorded as a corresponding abnormal evaluation type; And based on the preset first correspondence table, the abnormal deduction score corresponding to each abnormal evaluation type is identified to obtain the corresponding first deduction value; the first correspondence table is a correspondence table that reflects the correspondence between the abnormal type and the abnormal deduction value; the first correspondence table includes fourteen first correspondence records; the first correspondence record includes a first abnormal type field and a first deduction value field; the first abnormal type field includes the speed limit abnormal type, the U-turn abnormal type, the main and auxiliary road switching abnormal type, the stop and yield abnormal type, the first type of signal light abnormal type, the second type of signal light abnormal type, the third type of signal light abnormal type, the curve abnormal type, the ramp abnormal type, the tunnel abnormal type, the lifting pole abnormal type, the lane abnormal type, the unlighted intersection abnormal type and the roundabout abnormal type; the first deduction value field is an integer score; The sum of all the first deduction values ​​obtained is calculated to obtain a corresponding total deduction value; and the corresponding evaluation score for the time being is calculated based on the preset full score and the total deduction value = the full score - the total deduction value; and a corresponding first evaluation record is formed by each of the abnormal evaluation types and the corresponding first deduction values; and the corresponding first evaluation report is formed by all the first evaluation records obtained and the current evaluation score.

7. A device for executing the processing method for evaluating the perception response performance of a vehicle according to any one of claims 1 to 6, characterized in that: The device is connected to the vehicle-mounted system of the first vehicle; the device comprises: a data acquisition module, a data storage module, an abnormality identification module and an abnormality evaluation module; The data acquisition module is used to initialize two empty data acquisition queues as corresponding vehicle data queues and perception data queues and store them in the data storage module when the first vehicle starts the evaluation; and initialize fourteen empty exception queues to form corresponding fourteen types of exception queues and store them in the data storage module; the fourteen types of exception queues include speed limit exception queue, U-turn exception queue, main and auxiliary road switching exception queue, stop and yield exception queue, first type signal light exception queue, second type signal light exception queue, third type signal light exception queue, curve exception queue, ramp exception queue, tunnel exception queue, lifting pole exception queue, lane exception queue, no light intersection exception queue and roundabout exception queue; The data acquisition module is also used to collect data on the real-time motion, positioning and road information output by the vehicle-mounted system during the evaluation of the first vehicle, generate corresponding vehicle data records and add them to the vehicle data queue; and collect data on the target perception data output by the vehicle-mounted system to generate corresponding perception data records and add them to the perception data queue; The anomaly recognition module is used to perform fourteen types of perception response anomaly recognition processing according to the vehicle data queue, the perception data queue and the fourteen types of anomaly queues during the evaluation process of the first vehicle; the fourteen types of perception response anomalies include speed limit anomalies, U-turn anomalies, main and auxiliary road switching anomalies, stop and yield anomalies, type I traffic light anomalies, type II traffic light anomalies, type III traffic light anomalies, curve anomalies, ramp anomalies, tunnel anomalies, lift pole anomalies, lane anomalies, unlit intersection anomalies and roundabout anomalies; The abnormal evaluation module is used to obtain a corresponding first evaluation report by performing comprehensive scoring according to the fourteen types of abnormal queues after the first vehicle completes the evaluation.

8. An electronic device, characterized in that: include: memory, processors, and transceivers; The processor is used to couple with the memory, read and execute instructions in the memory, so as to implement the method according to any one of claims 1 to 6; The transceiver is coupled to the processor, and the processor controls the transceiver to send and receive messages.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a computer, the computer executes the method according to any one of claims 1 to 6.

Images